My journey of improving the AI

4X strategy game from Proxy Studios

Moderators: Pandora Moderators, Slitherine Core

JetJaguar
Lance Corporal - SdKfz 222
Lance Corporal - SdKfz 222
Posts: 23
Joined: Mon May 13, 2013 4:21 pm

Re: My journey of improving the AI

Post by JetJaguar » Sun Jan 18, 2015 7:14 pm

Xilmi wrote: And here's what I think about diplomacy:

Bribing AIs to go to war with each other seems a big issue. Due to this I've turned certain doom into victory many times.
Also bribing someone to go to war with someone you aren't at war with yourself is even worse. Preemtively getting them off your back and have them devote their economy to military while you can savely tech up and build infrastructure is a cheap trick to dramatically lower difficulty.
Two AI's that are both stronger than you? Let them kill their units on each other and then join in with the winning side to snatch some cities!
AI should be a lot more wary about this stuff. It should still be possible but way harder.
Then everything that involves paying credits in diplomacy is just a joke right now. Why? Because the amount of credits asked for is relative to what you have. For the AI it doesn't make a difference if you pay 500 credits or 22, if that's what you have. So just by spending your credits or running low taxes to not generate a lot of them in the first place, you will basically make tributes and payments to have them sign contracts a non-issue.
The amount of credits asked for should scale with what you could have! You want that contract signed? Then save up for it! Not spend your cash on a buyout and get the contract signed for your remaining 16 credits.
Glad you're on this Ail because I think this is major problem. Can the gift/bribe credits be modded to a fixed amount?

also: in the meantime, what would you suggest for a 'house rule" to deal with this issue? maybe only allow yourself to give credits when you already have 200 credits in your treasury?

Thanks in advance.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Sun Jan 18, 2015 8:01 pm

Well, yesterday I had an intense look at diplomacy. I decided to push it to later and do some other things first. (City-placement and border-expansion it was today)

The reason is, that it is really huge and complex and quite rework-worthy. Another problem is, that it is much harder to test because in observer-mode you don't get to see much about diplomacy.

A house-rule would be to never suggest anything on your own and just react to what the AI asks for.

However, if you're playing 1.5.5-beta on higher difficulty-levels you might need any exploit you can come up with anyways. ^^

JetJaguar
Lance Corporal - SdKfz 222
Lance Corporal - SdKfz 222
Posts: 23
Joined: Mon May 13, 2013 4:21 pm

Re: My journey of improving the AI

Post by JetJaguar » Mon Jan 19, 2015 3:05 am

Xilmi wrote:A house-rule would be to never suggest anything on your own and just react to what the AI asks for.
I'm thinking that maybe I could allow myself to actively attempt to pay/bribe the AI, but only if I have a significant amount of credits in my treasury. I would have to, as your other post suggested, have the incentive to save up enough credits to start parlaying with the AI. Now I'm just trying to determine the exact amount of credits. I suppose that number should increase as the game progresses. What do you think?

Thanks again for all your work on the AI. I'm thoroughly enjoying your 1.5.5-beta.

Evrett
Corporal - 5 cm Pak 38
Corporal - 5 cm Pak 38
Posts: 41
Joined: Mon Mar 18, 2013 6:30 pm

Re: My journey of improving the AI

Post by Evrett » Mon Jan 19, 2015 6:42 am

This is really exciting Xilmi! Its great to see someone actually care and get involved! Thank you for your service! If you solve the ai jumping units into the water and the Ai not managing unit health please please share your work with the game Warlock2: the Exiled. A great fantasy 4x whos AI suffers from these same issues. It makes me wonder if these smaller developers buy some sort of wholesale AI template for their games rather than writing the code themselves.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Mon Jan 19, 2015 8:55 am

@JetJaguar: Maybe Population * 5 would be an appropriate scaling-value.

@Evrett: Units jumping into the water and health-management is something I took a look at on saturday. So you can expect an improvement for whenever the next beta-patch hits.
No, I can guarantee you that the AI-code of Pandora is not based on any template. It is closely wired with the rest of the code which is very specific.
There's no way to carry anything over to other games. And even if it was, doing so would clearly violate the NDA I signed.
What I can do is encouraging Warlock2-Enthusiasts who happen to be programmers as well to go ahead, try the same thing and ask the Devs if they can get involved. They might actually agree. ^^

SunDevil
Private First Class - Opel Blitz
Private First Class - Opel Blitz
Posts: 1
Joined: Tue Jan 20, 2015 11:41 pm

Re: My journey of improving the AI

Post by SunDevil » Tue Jan 20, 2015 11:44 pm

Xilmi,

Thank you for all your work. If I'm using your AI Mod for Nashira, is that going to work with the beta patch? Is your mod now obsolete?

Thanks.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Wed Jan 21, 2015 12:59 am

Yes, the mod is obsolete with 1.5.5.

Sawelios
Lance Corporal - Panzer IA
Lance Corporal - Panzer IA
Posts: 12
Joined: Mon Jan 05, 2015 12:02 pm

Re: My journey of improving the AI

Post by Sawelios » Wed Jan 21, 2015 12:42 pm

Hey, guys, are there any sculptors around? We owe a monument to Xilmi.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Mon Jan 26, 2015 12:42 pm

With the second version of 1.5.5 being out, here's my agenda for the next one:

Revisiting miner-allocation:

The current mechanism results in switching the use of miners back and forth.
While it still somehow works, it has negative side-effects on tile-improvement selection and most likely is not ideal anyways.
I want to rewrite it so the mining allocation is more stable.
It should become closer to human behavior, where the amount of miners often is adjusted to efficiency of the sources but without striclty limiting it either.

Revisiting tile-improvement-selection:

Right now it is tied too much to the current situation in the city. Especially when it comes to the importance of mines and pollution-processors.
Switching population-assignments back and forth thus also results in wanting to switch tile-improvements.
This can lead to effects where formers get stuck constantly switching the improvement of a tile.
The need of tile-improvements should be evaluated more on general strategy rather than the current situation.
In addition there should be a mechanism, that prevents changing an existing improvement as long as there are completely unimproved tiles nearby.

Strengthening city-defenses:

I already implemented some improvements here. They work well for cities that actually build units.
However, the request for defenses is diminished by the distance a unit has to travel in order to do so.
I think this should be changed and therefore every city should be well defended.
Also right now the minimum request of units scales with what was seen from others. This can be dangerous early on, when a city doesn't build any unit before encountering an alien.

Revesiting technology-choices:

The current mechanism works well for must have-techs but not for situationoal ones. There needs to be a bunch of mechanisms that for example make sure that at least one current-era unit template is available without tagging all current-era-unit templates as must have. Same for weapons. e.G.: They all have high initial value but when you get one of them, the rest loses part of its attractivity.

And a long term goal for the future:

Fewer but better prepared wars:

Currently the choice of going to war is based on standings, aggressiveness and a comparison of military-power. If these align in a certain way, war will be declared. Good players, however, plan, prepare and execute wars always with a certain goal in mind.
For example capturing a certain city.
Also they usually base the war on a stable economy. For example: Starting or even preparing a war, when you have morale-problems at home usually is not a good idea.
Also: If they realize the goal is not achievable, they will not just continue the war for no good reason.

Players that are declared war on, usually do not have a goal, their aim is to mainly hold their ground and defend as good as they can.

So what I want to do is the following: Before even thinking about declaring a war, the AI has to make sure it's economy is in tip-top-shape and it is not suffering from negative morale.

If it feels, it can afford a war, it first needs to picks a target, usually a certain city. It then will try and get reconnaicance information about that city. Units that are good at fighting at the target-location should be prioritized: e.G: Ships, if it's a coastal city, Troopers, when the city is mostly surrounded by rough-terrain and mechanical units, when the city is mostly surrounded by plains.
Operations like WMDs should be saved up for pressuring the target. War declaration only happens when the units are brought in position to attack.

Someone, who gets declared war on, shall not do the same thing as the attacker. Instead he shall keep his units at home and defend. Only when there is no combat happening on his territory, he can then go into the preparation-phase himself and see if he can find a suitable war-goal.

An AI, that already is at a war should refrain from declaring another one.

Zak0r
Sergeant - Panzer IIC
Sergeant - Panzer IIC
Posts: 188
Joined: Sat Jan 25, 2014 1:21 pm

Re: My journey of improving the AI

Post by Zak0r » Mon Jan 26, 2015 3:43 pm

That sounds really cool. Especially the paragraphs about actually having the AI planning a war and not only sending units and see what happens.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Mon Jan 26, 2015 6:26 pm

Zak0r wrote:That sounds really cool. Especially the paragraphs about actually having the AI planning a war and not only sending units and see what happens.
Yes, that one was greatly inspired by our talks during the last game.
When you were wondering why I didn't use my window of opportunity to attack you, I realized, that it mostly was because I didn't even know that this opportunity was there and I haven't even though about any plan of what exactly to do.
That's why my 100-200 Str-units just idled in the cities. ^^
Then you spoke about how you would have done it, if you where me and it was quite inspiring.

But this clearly is a very ambitious task, much more so than some pop-allocation-algorithm-rework.

But some things about it are easy. Like the: Don't declare war while you have morale-problems-bit.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Tue Jan 27, 2015 1:44 pm

Xilmi wrote: - Revisiting miner-allocation
- Revisiting tile-improvement-selection
Both of these have been done yesterday and the results look a lot better than before.
Population-allocation now also takes pollution into account to some degree. Higher pollution will increase the value of scientists as they produce less of it.

I'm still experimenting with the city-defenses.
The problem is not to have them use more units for this but how those units should react, when the city is actually under siege.
If they walk out, they can be killed much easier and then the city is, once again, is easier to caputre. If they stay in, the tiles around the city can easily be pillaged and they could be softened up by constant AoE-attacks, which they cannot heal agaist.
It has to be decided on a case-to case-basis.
Maybe stay inside, if the units lingering around the city are stronger and come out if they are weaker.

Also right now exploring the world doesn't work for some reason. I have to find out why. AI has to make contacts and try and gather goodies too.
On the pro-side: not even bothering with exploration keeps their early unit losses at a very low level.

Zak0r
Sergeant - Panzer IIC
Sergeant - Panzer IIC
Posts: 188
Joined: Sat Jan 25, 2014 1:21 pm

Re: My journey of improving the AI

Post by Zak0r » Wed Jan 28, 2015 7:45 am

Xilmi wrote: It has to be decided on a case-to case-basis.
Maybe stay inside, if the units lingering around the city are stronger and come out if they are weaker.
That's more or less what a human player would do so I think you're on the right path here.
Xilmi wrote: Also right now exploring the world doesn't work for some reason. I have to find out why. AI has to make contacts and try and gather goodies too.
On the pro-side: not even bothering with exploration keeps their early unit losses at a very low level.
I usually explore with weak units until the alien aggression starts. The next wave of exploration starts when the explorer can easily manage the xenomorphs or with ships (that can easily manage the carcharodons).

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Wed Jan 28, 2015 12:54 pm

Zak0r wrote: That's more or less what a human player would do so I think you're on the right path here.
I have implemented it in this way. But it needs to be properly tested.
The come out when enemy is weaker-part works well, as can often be seen against aliens.
However, in the test-games a city was rarely attacked by enough forces to trigger the "stay inside"-part.
The reason is: We have the uncoordinated attacking from before but with less units, since everyone keeps more units to defend.
The attackers often-times didn't even dare to walk into the cities radius.
Captures only occured to off-site-cities far away from the defenders home-terriotry or later on with black-holes.
I have started a game where I play myself once again. Economically the improvements of worker-allocation and the bugfix to tile-improvement seem to work well. I'm quite far behind in everything (playing on Hard difficulty).
But usually it was the later stages of the game where I could catch up and pull ahead, so I'll have to see how it goes this time.
Zak0r wrote:
Xilmi wrote: Also right now exploring the world doesn't work for some reason. I have to find out why. AI has to make contacts and try and gather goodies too.
On the pro-side: not even bothering with exploration keeps their early unit losses at a very low level.
I usually explore with weak units until the alien aggression starts. The next wave of exploration starts when the explorer can easily manage the xenomorphs or with ships (that can easily manage the carcharodons).
The reason why it didn't explore was a combination of several issues.
The most important one was that the exploration-agent generates a maximum of three targets to explore. But then these targets where checked for their viability. But I had forbidden land units to explore water. So when the top-three priorities where water-tiles it would never explore unless it had ships.
Now I have no longer forbidden it but instead just decrease the value of water-exploration.

Another issue was that with more units assigned to defending, there were less left to do exploring. And without exploring, the AI never saw a reason to make more units until the aliens actually came into the city-radius. And then it prioritized fighting aliens over exploring. I reduced the required defender-count for when an AI had not made contact yet to 1 per defendable location.
I also experimented with reducing it to zero before alien-agression starts, so it would explore with it's starting unit. However, this had resulted in the second city being lost to alien and it seems too dangerous, if another faction finds you first and takes your capital right away.

However, the issue of not taking ruins should be resolved by it and contacts are also made much quicker.

All of that helped me to better understand how units are managed. That first was a big mystery to me but now it becomes clearer and clearer.
Now I finally feel competent enough to making the AIs smarter about how to use their spies. I tried before but had very strange effects. Most likely because I didn't understand how it works.
Did you know that you can click a spy-option and then the target-city even if the spy is not at the city yet? I learned that through how the AI does it. ^^
And if this works, the next step would be implementing better tactics like flanking and moving to tiles that grant an attack-bonus before attacking.

JetJaguar
Lance Corporal - SdKfz 222
Lance Corporal - SdKfz 222
Posts: 23
Joined: Mon May 13, 2013 4:21 pm

Re: My journey of improving the AI

Post by JetJaguar » Wed Jan 28, 2015 4:46 pm

I have started a game where I play myself once again. Economically the improvements of worker-allocation and the bugfix to tile-improvement seem to work well. I'm quite far behind in everything (playing on Hard difficulty).
But usually it was the later stages of the game where I could catch up and pull ahead, so I'll have to see how it goes this time.
Ail, do you usually play on Hard difficulty? How exactly do you set your other pre-game set-up options?

Thanks in advance.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Wed Jan 28, 2015 10:18 pm

Before I started programming on the game, I played and usually won on very-hard.
For my test-games I try different settings.

I was kind of hoping, that hard would be hard enough but unfortunately it wasn't.
One problem is that the anti-runaway-mechanism can also help you.
Terra-salvum where the runaways and so I just had to be opportunistic and start attacking them when the others did so too. In the end I got the edge in science and everyone hated me.
Couldn't really make progress against the three remaining AIs but holding my ground is easy, since they don't have black-holes yet.

When fighting against terra-salvum, I had to actually crack some tough nuts. But the plans i had to come up with will eventually be utilized to teach the ai.
So how to conquer a city with strong defenses?
Surround it as good as possible for double-flank and to prevent troops from somewhere else getting in. Kill everything that comes to help.
Use exactly one operation (Nuke/Bombard) on it. This prevents healing and since you only use one of them each turn, you'll not run out. Eventually they become weak enough to take them out.

So a worthwhile thing to improve is how operations are used. Seeing several nukes wasted on a sole 100-strenght-unit just hurts to watch.

Also one thing that I noticed in particular is how many wars are started via declare war on third party. AIs use that on each other all the time and thus get into wars that normally would never have occured. Someone got preston to declare war on me several times despite him being friendly towards me (and generous towards who asked him).
Easy counter: Just ask someone who is generous towards me to declare war on him aswell. Extremely cheap technique.
The result of the request should not be affected by the relations you have when asking, it should only work when you would want to declare war on your own.

JetJaguar
Lance Corporal - SdKfz 222
Lance Corporal - SdKfz 222
Posts: 23
Joined: Mon May 13, 2013 4:21 pm

Re: My journey of improving the AI

Post by JetJaguar » Wed Jan 28, 2015 11:10 pm

Xilmi wrote:Easy counter: Just ask someone who is generous towards me to declare war on him aswell. Extremely cheap technique.
The result of the request should not be affected by the relations you have when asking, it should only work when you would want to declare war on your own.
It does seem like almost everytime I ask some-one who's generous (sometimes friendly) towards me to declare war on someone they don't like, it's automatic. Often this happens on the same turn I receive the notification that they just made peace. I guess there should be a minimum peace treaty duration. Going to war with another faction on the very same turn you made peace (or even a few turns after) feels wrong.

I'm glad that this part of diplomacy is on your list. I'm especially glad that you're changing the AI to only declare war when it's prepared for one and has specific goals. For most factions (the exception maybe being The Divine Ascension... though not defininately) just being angry or furious at another faction shouldn't necessarily mean war. It should result in not making economic or research treaties, denouncements, demand... ect.

It will no longer be so easy to take advantage of the AI when it's sent it's few units off to fight in some far-off war because now it will only declare war when/if it's ready.

Again, thanks so much for all of your efforts. In my opinion, you're prioritizing all the right things.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Thu Jan 29, 2015 11:59 am

JetJaguar wrote:
Xilmi wrote:Easy counter: Just ask someone who is generous towards me to declare war on him aswell. Extremely cheap technique.
The result of the request should not be affected by the relations you have when asking, it should only work when you would want to declare war on your own.
It does seem like almost everytime I ask some-one who's generous (sometimes friendly) towards me to declare war on someone they don't like, it's automatic.
Exactly. I fixed that after my post. First of all: The war-viability-check I had programmed earlier was not called in this scenario and secondly there was a bonus added to the war-declaration-likelyness if somone asked you. I removed that bonus.
JetJaguar wrote:Often this happens on the same turn I receive the notification that they just made peace. I guess there should be a minimum peace treaty duration. Going to war with another faction on the very same turn you made peace (or even a few turns after) feels wrong.
I agree I was looking for an easy fix for that but it is a little more complicated. I basically would have to replicate the Non-agression-pact-mechanism for that. But I'll keep it in mind for later. The root cause of the problem should be fixed anyways.
JetJaguar wrote: I'm glad that this part of diplomacy is on your list. I'm especially glad that you're changing the AI to only declare war when it's prepared for one and has specific goals.
Yes, it is because I identified their wicked diplomatic behaviour to be one of their biggest remaining weaknesses. I know that their economical-buildup mechanisms like population-allocation and tile-improvement-selection are quite decent now. Their build-order could need a little work still but it's not too far off.
An interesting sight was that The Ambassadors usually were doing well in every single game. Simple reason: They usually stood out of wars and thus could focus on economy.

The declare war only when prepared is next on my agenda. Should be relatively easy. I just need to count what is left after assigning all defenders and attackers then compare it to what it thinks is needed to take a city of who they otherwise would immediately declare war on.
A positive side effect will be that the existence of nearby aliens will automatically reduce the chance of declaring war since they will assign units to attack those too, which takes away from their "what is left".
I'm really interested to see how that turns out.

And just a reminder to myself: The mechanism of reducing standing against aggressors also needs to be implemented.
JetJaguar wrote: For most factions (the exception maybe being The Divine Ascension... though not defininately) just being angry or furious at another faction shouldn't necessarily mean war. It should result in not making economic or research treaties, denouncements, demand... ect.
It will no longer be so easy to take advantage of the AI when it's sent it's few units off to fight in some far-off war because now it will only declare war when/if it's ready.
I have the exact same opinion on that. Bad relationship and war should really be regarded seperatly. War only makes sense if it is likely to be benefical for the aggressor.

JetJaguar
Lance Corporal - SdKfz 222
Lance Corporal - SdKfz 222
Posts: 23
Joined: Mon May 13, 2013 4:21 pm

Re: My journey of improving the AI

Post by JetJaguar » Thu Jan 29, 2015 10:56 pm

An interesting sight was that The Ambassadors usually were doing well in every single game. Simple reason: They usually stood out of wars and thus could focus on economy.
I have noticed this as well and believe that it's very telling. The other factions are wasting their production on wars that don't seem to result in successful conquest very often (I'm usually the only one who's conquering cities, not the AI). When you change this, I think the result will truly be a quantum leap forward. As the player it's so easy (and exploitive) to manipulate this. I think that my easy conquest is almost always the result of the AI faction's lost units from constant warring. I suspect that the AI is sending out units a little at a time and that they are getting picked-off "piece-meal" at the city walls of other AIs; and so are totally wasted.

With the only real point of offensive warring being to conquer cities, the AI is at a tremendous disadvantage here compared to the player.

I do have an exploitive habit of taking cities from the AI when another AI that they're at war with has destroyed most of a city's garrison units (and are one turn away from taking it). It's easy to say: just make a "house rule" not to do that, but sometimes we both share in the destruction of the garrison, so who should have the rewards? This is something the player will always be superior at. I sometimes wish there was a mechanic in place which made the AI angry at you for robbing it of it's conquest at the last minute; but the aggression towards conquerors that you are adding should cover this to some degree. With that in place, the player (and the AI, for that matter) should always ask himself if the conquest of a city is worth the potential diplomatic repercussions. I'm really looking forward to that. I think it will be a real game-changer.
have the exact same opinion on that. Bad relationship and war should really be regarded seperatly. War only makes sense if it is likely to be benefical for the aggressor.
I think that really bad relations ("furious") should be enough to motivate war; but motivation without preparation is futile. Basically, I guess the AI should need to have both (bad relationship and prepared military) to make a declaration of war. Poor relations with an aggressive faction should also make it prepare for war defensively as well.

also: I notice other factions (most often, The Ambassadors) contacting me and requesting/demanding that I make peace with a third party. If refused, do you suffer a relations hit with the faction who asked you to make peace with a third party? If so, I think that these request should happen more often. It's a perfect way to have a warning system in place; after all, they don't want to be hostile with you for conquering cities.... they'd rather you just not conquer cities in the first place -for balance of power purposes (preventing runaways). It's also good for the conqueror to know that he has reach the line and is being dared not to cross it. without having crossed it already.

Good luck with this. I'm eagerly awaiting playing with this AI that is keeping an eye on conquerors. How far along are you? Thanks in advance.

Xilmi
Pandora Community Developer
Pandora Community Developer
Posts: 520
Joined: Fri May 23, 2014 3:21 pm

Re: My journey of improving the AI

Post by Xilmi » Fri Jan 30, 2015 2:22 pm

JetJaguar wrote:I suspect that the AI is sending out units a little at a time and that they are getting picked-off "piece-meal" at the city walls of other AIs; and so are totally wasted.
Nope, they were not that stupid. ^^ They mostly strolled around the cities and eventually there were fair and even fights, where the defender would come out and they'd battle. So losses on both sides where relatively equal.
Did you know that in the advanced-game-settings you can, instead of picking a faction, set yourself to observer and just watch what the AI's are doing?
That's how most of my testing is done. I can see all the dumb things they do and then then see what I can do about it.
JetJaguar wrote:How far along are you? Thanks in advance.
There won't be a patch this weekend. So most likely around next Friday.

The reason, however, is not that I've been lazy. More like I'm redoing huge interconnected parts.
The planning and preparation-part already partially works.
Wars are rare now, but when they happen, captures are quite likely.
I also fixed an issue that troops that can't take a city just did nothing. They now will pillage the crap out of the surrounding tile-improvements. (before they mostly only pillaged on tiles 3 tiles away)
Operation-usage has been a big part of what I've done. AI will now rather nuke your observatory instead of your 100 strength-tank. And good luck with using transports. They are now considered as valuable targets as they are.
I did some more adjustments to population-assignment and probably will do even more. I think new cities should use more workers and less scientists, while it should be the other way around in better-developed cities. That is also part of why I think the AI is still struggling to keep up economically. (New cities develop too slowly and there's too little science produced in comparison since the infrastructure is not in place yet).

Please let me know about all the house-rules you are considering or using because of AI-weaknesses. I really don't want you to have to. I want everything exploitable to be resolved. ;)

The new way how I want relationship to work is that it plays a role in picking the next target, whereas targets are cities. I'm calculating a value for all known cities.
The value is affected by the cities size, how well it is defended, how far it is away and then I multiply it's value with (1-Standing towards it's owner) whereas standing is a value from 0 to 1. So Furious means the city is fully valued as a target while generous means that your cities are not considered as targets at all, even if it otherwise looks juicy.
War then only is declared at the one who owns the city that is desired most (but always keeping in mind, that bad relations mean higher desire). Additionally it had to be made sure that there are enough troops available to assign to this task and that there are no morale issues.
The no morale-issues part will also lead to difficulty-level having an effect on the likelyness of war. On lower difficulties the AI might struggle with morale, whereas they usually shouldn't on higher ones.

Post Reply

Return to “Pandora - First Contact”