A Better AI.

also is there any way to tell this mod has loaded once your in the game, (other than the different AI behaviour)

Hold down ALT hover mouse over your own name in the score list, should say <<Better AI>> or something like.
 
I can assure you we are careful about balance. There is no question that as the AI behavior changes, becoming 'smarter' it is going to make it harder to get away with some things which were easy in the default game. If you like to play at or near deity level, winning by exploiting the stupid behavior of the AI, then this mod might not be for you. The whole point of this mod is to make the AI behave in a smarter fashion, so that you can go to an easier difficulty level, where the AI gets less of a handicap advantage, and still have a challenging game.

Back to tech trading: if it is harder to exploit the stupidity of AIs to fund your research while gaining nothing out of it, that is a good thing. If tech trading becomes completely removed, that is a bad thing. I believe that we are falling somewhere in the middle, but if that is not the case, I would like to see specific examples.

-Iustus


You intentionally missing my main point, insted of accusing some one of "Wining by exploiting AI weekness". Think about argument!

I probably should just stop waster my time, as you definatly not interesting in analising consiquences of your changes.

Yes, you can spend money to buy tech, but you will pay 2 (I think) times or more then it will take you to self research it. It is not economically valible.

I undestand that it is a good idea to make AI spend and use money they have.
I am argue, that this should increase VALUE AI PUT on it. Having money should become more valible to AI, as it is for human.

May be not up to old level, but some increase in value AI put on money in trade should accure.

Otherwize trading for money had no meaning.
 
You intentionally missing my main point, insted of accusing some one of "Wining by exploiting AI weekness". Think about argument!

I probably should just stop waster my time, as you definatly not interesting in analising consiquences of your changes.

Yes, you can spend money to buy tech, but you will pay 2 (I think) times or more then it will take you to self research it. It is not economically valible.

I undestand that it is a good idea to make AI spend and use money they have.
I am argue, that this should increase VALUE AI PUT on it. Having money should become more valible to AI, as it is for human.

May be not up to old level, but some increase in value AI put on money in trade should accure.

Otherwize trading for money had no meaning.

Did you make the test he talked about? About a trade of a tech for money with and without Better AI dll in the same game/turn?
 
Hold down ALT hover mouse over your own name in the score list, should say <<Better AI>> or something like.

Iustus,

This seems to be a FAQ. Maybe you could get Blake to edit the original post, or put it in you signature or something?
 
You intentionally missing my main point, insted of accusing some one of "Wining by exploiting AI weekness". Think about argument!

I probably should just stop waster my time, as you definatly not interesting in analising consiquences of your changes.

Yes, you can spend money to buy tech, but you will pay 2 (I think) times or more then it will take you to self research it. It is not economically valible.

I undestand that it is a good idea to make AI spend and use money they have.
I am argue, that this should increase VALUE AI PUT on it. Having money should become more valible to AI, as it is for human.

May be not up to old level, but some increase in value AI put on money in trade should accure.

Otherwize trading for money had no meaning.

Mutineer,

I understand why you would be concerned, since portions of this mod have made their way into a patch released by firaxis. I cannot tell you just to ignore this mod if you don't like it.

If you don't like it, then help make it better.
1) Give a savegame and concrete examples of the problem you think exists. Iustus explained how you can do this. Your latest claim is that the human player will have to pay twice as much for a tech than before. This should be easy to verify.
2) Come up with a valid reason why an AI goal should be to have a large balance of funds, other than to give money to the human player. I can think of one, to upgrade units easier. I'm sure you can think of others.

In short, please stop using generalities in your post. It gives the appearance that you haven't downloaded and tested the mod and that you are only raising objections based on posts and release notes. You have downloaded and tested the mod haven't you?

You are obviously an experienced player who cares about the game. I (and many others including Firaxis) think this mod is a great step forward for the AI of the game. Your opinion is important, but you need to back it up with some examples.
 
1) Give a savegame and concrete examples of the problem you think exists. Iustus explained how you can do this. Your latest claim is that the human player will have to pay twice as much for a tech than before. This should be easy to verify.
To be fair(and while I'm not sure he's right and he sure likes to write in an inflammatory way), that is not is claim, and what he says might be valid.

What Mutineer is saying is that while the value in gold of a tech to the AI is indeed the same with or without BetterAI, it should not be. Before BetterAI, gold had little to no value at all to the AI. He basically only accumulated it and rarely used it, so there was no reason for him to be interested in trading for more. Now that the AI actually uses its gold, shouldn't he place a higher value on it, thus asking for less gold than before when trading for a tech? It seems that's what the human players would be doing, since they know they can use this gold to fund deficit research which will help them tech further in the long run.

So the point I see in between the flaming in Mutineer's posts is that it's not everything to say "we did not change this or that"(in this case the value of gold to the AI), but it might be important to go further and say "should we now change this or that to make things more balanced/preserve this aspect of play". Whether he is right in the specific example he gave here, I'm not good enough at Civ4 to judge.
 
Maybe I did mis-state his claim.

you will pay 2 (I think) times or more then it will take you to self research it.

This should still be easily verifiable with a simple case study, and it should be incumbent on him to do that, not the Better AI team.
 
Maybe I did mis-state his claim.



This should still be easily verifiable with a simple case study, and it should be incumbent on him to do that, not the Better AI team.
You mis-read his claim is all :)...

It says you pay twice more money to the cpu for a tech that it would cost you to research it. Never in any of his posts does he stats that you pay more than before BetterAI.

In fact if you re-read post 851 in this thread, he specifically says that Firaxis made it so that "buying tach for money stop to be a valid startegy and become economically bad". He then goes on to state that the changes in BetterAI coupled with that fact creates a bad situation. Never does he claim that BetterAI raised the cost to trade for a tech.

I know his english is bad(clearly not his first language so I don't think we can blame him for that), but it's still readable :)...
 
In civ3, it was often cheaper to buy a technology than to research it. As an additional advantage, buying allowed you to time the acquisition of a technology so that you could directly trade it to other AI's and get more money back than what you had spend on buying it in the first place.
Players therefore didn't invest in technology and technology improving buildings like libraries and universities. They were only interested in money and money improving buildings. Money allowed them to research, to upgrade units and to rush buy stuff and it helped them pay the maintenance cost. And thus a whole section of the game related to technology research was lost.

This was a tactic mainly used by the more experienced players or those who read about it in the War Academy.

In civ4, this situation was avoided by two things. In the first place, buying technology was made more expensive than researching it yourself. Next, they put a limit on the number of technologies that could be acquired by trading. The limit is related to the relation with the AI and the personality of the AI. So now you'll have to do some technology research yourself and both gold and science are valuable commodities. Gold is needed for all kinds of upkeep and for upgrading units and rush buying and spying missions. Sometimes gold can be used in trade. Science is of course important to invent new things.

You could argue that the lesser availability of gold in the AI hands in the BetterAI mod (because it now knows how to use it in deficit research) has made gold less available in trading. Thus it makes gold a lesser influence in the technology research/trading. Whether that is a good thing or a bad thing is a thing of personal taste.


I don't know how much more it costs to buy a technology than to research it yourself. I've noticed that the cost of a technology in trade can vary a lot. Especially the miliatary technologies can be rather expensive to buy compared to researching them yourself. Personally, I think that is a good thing as a human wouldn't easily trade away a military technology. A human would think about that hard before trusting an AI with such dangerous technology.

I personally have no problems with it, but I think this is the issue which Mutineer is trying to explain while not showing a lot of good taste in wording it. A bit more tact could be useful to get people to listen to you.
 
Finally had the time to test a full game of BetterAI (Dec 9 build, Monarch).

Nice work, although I have two comments:

a) A better defensive AI would be nice. At the moment there a few units in each city and spordic attacks by single offensive units, which of course doesn't work. I also think any civ, even a builder one, which loses a city should go into "panic mode" and build/emphasize militay for a period of time.

b) City placement. Somehow the suggested (blue) locations for settlers is never where I build cities. And the AI cities always seem in 'strange' locations. I cannot quantify it, but AI cities seem too close to each other.
 
However I now what to use this mod. Where do I put the .dll for LAN mulitplayer, I tried just putting it straight into the Assets (not custom assets) folder but this crashed the game before it even loaded.

Hmm, are you 100&#37; sure you were using the right version, warlords vs vanilla civ? The only time I have seen crashing is when someone is trying to use a vanilla dll with warlords, or a warlords dll with vanilla civ. Or possibly you have warlords, but have not updated to 2.08 (I am not sure if that crashes or not).

Mulitplayer LAN seems to ignore the custom assets when I try to start a game, as well as ignoring it if I try to load it as a mod.

I thought it was supposed to totally ignore custom assets when playing multiplayer, but that is not the behavior I experienced when Blake and I tested multiplayer the other day. I suppose this could be because we both had cheat code set in my CivilizationIV.ini file (in My Games:Warlords):
Code:
; Move along
CheatCode = chipotle

I did not think that was it, befcause in multiplayer, cheat code is not active, you do not see the extra info, and you cannot switch to god mod.

So, when Blake and I tested, we both had the BetterAI dll in the custom assets folder and multiplayer worked.

Hopefully, someone more experienced with multiplayer can speak to this issue.

also is there any way to tell this mod has loaded once your in the game, (other than the different AI behaviour)

Hold down the alt key and mouse over your name in the list of player scores in the bottom right side of the screen. It will say "=== Better AI === (build date)"

-Iustus
 
Hmm, I saw that there was a new build (dated 12-12). I triied using it but it generates a CTD immediately upon finishing world creation etc. My guess is something in the new starts might be buggered up?

Anyone else see this or is something that I'm doing?

Thanx!
 
First post in this thread. WOW!
Played my first game with Better AI loaded as a mod. I got crushed on Monarch at about 1000AD by Alexander. He showed up with 2 stacks that were well balanced and contained about 20 units each. He was ahead in techs as well so the stacks were not out dated. Obviously, the AI is better, my question comes about city placement. Three of the civs that I could actually see the capitols of (Alex, HC, Cyrus) had all placed cities directly on top of resources and had 3 - 5 bonus resources in the fat cross. HC actually had copper under Cuzco. I on the other hand had non irrigable rice as my only resource in my fat cross (random start settings with no map regeneration). Is this a "normal" situation on this difficulty level? I will obviously move down to prince and possibly regenerate the map until I see something I like on my next attempt. Cudos for the fine work! How long until Firaxis pays you to add this to their base product?!

Reading back over the earlier posts, I see reference to the "Better AI" being included in a Firaxis patch, could someone please enlighten me as to which one?
 
You intentionally missing my main point, insted of accusing some one of "Wining by exploiting AI weekness". Think about argument!

I probably should just stop waster my time, as you definatly not interesting in analising consiquences of your changes.

Yes, you can spend money to buy tech, but you will pay 2 (I think) times or more then it will take you to self research it. It is not economically valible.

I undestand that it is a good idea to make AI spend and use money they have.
I am argue, that this should increase VALUE AI PUT on it. Having money should become more valible to AI, as it is for human.

May be not up to old level, but some increase in value AI put on money in trade should accure.

Otherwize trading for money had no meaning.

Mutineer, kindly lose the attitude.

I can assure you that if changes result in the AI losing personality, that is basically considered a bug to be remedied. No-one is more fanatical about AI personality than I (others may be equally fanatical ;)). As examples of this in the most recent build you'll see that Theocracy running AI's are less likely to pursue cultural victory - instead they are more likely to use the "Missionary" strategy which is more steadfast devotion to their state religion (especially one they founded), I feel this much better fits the character of Saladin and Isabella.

As far as money issues goes. One opinion is that the AI should never give up money (held by me and seemingly tofof a couple of pages back).

Basically, balance-wise, it's best that AI's don't give up large amounts of cash. Then again it's nice to have a small amount for "war fund donations" (that's pleading for some cash to upgrade units) and making change in tech trades.

If I had my way the AI would trade for cash under two conditions:
1) When they are leading in tech they'd sell their outdated techs to anyone stupid enough to buy them, further cementing their tech lead due to faster research (I'd rather the AI's be wise to this trick, and instead trade amongst themselves and shun the tech leader)
2) The AI's should help fund the war efforts of allies who are at war with a mutually disliked foe.

Mutineer, I am uncertain whether you want it to return to 1.61 behavior (AI always trades away surplus gold) or whether you want us to give the AI some solid logic for valuing money, your accusatory attitude suggests that you think we've broken something which was working fine but I may be just reading it that way. To the former, I can say "Hell no" - the AI will continue to spend their cash rather than coughing it up for outdated and useless techs to further accelerate tech pace. To the latter I'd have to say that it may be outside the scope of the project - it is certainly a controversial issue and with those it's safer to leave it with the "broken, but developer endorsed" state, rather than risking a "still broken, no longer developer endorsed" state. If I can think of a really good all-encompassing way for valuing gold then something might happen, but for now I feel that simply being stingy with their gold is closer to "good balance" that the 1.61 state which was definitely unbalanced.

So far as Tofof's concerns of the AI paying too much for resources... I've kind of noticed this. I wouldn't underestimate the use they may be able to get of the resource now and it's kind of a tradition that the AI will make 1 for 1 trades in CIV. I'm probably not going to change it but I may look at reducing the maximum amount they'll pay.
 
Back on subject....

Blake & Iustus, I saw an interesting thing in my current game. Isabella was Confucious (Hatty was next door between me and Izzy, and had founded confucianism). Then, Izzy founded Christianity. Great. But, then I continue to see Spanish confucian missionaries stream by (along with Egyptian ones), on their way to convert Huyana (on the other side of me).

I would have thought that Izzy would change to her "home grown" religion, rather than continue to propagate the "foreign" one.

Yeah, I know, she is only acting the way she's programmed. I guess that's my point... :D Anyway, something you might want to consider.

Wodan
 
So far as Tofof's concerns of the AI paying too much for resources... I've kind of noticed this. I wouldn't underestimate the use they may be able to get of the resource now and it's kind of a tradition that the AI will make 1 for 1 trades in CIV. I'm probably not going to change it but I may look at reducing the maximum amount they'll pay.

I can see why Tofof is concerned about the issue. If you have many surplus resources and can do nothing better with them than get some gold for them, then you can get quite an amount of gold. It can sometimes change the gold rate needed for maintaining your civilization by 10&#37;.

However, if you want to buy a resource from the AI for gold per turn (because you don't have many resources and need some for happiness or health), then you'll have to pay a lot more for these resources.
Typically I have to pay 30 or 40 gold per turn in the 0-500AD time period for a happiness resource (huge map, emperor, continents). The AI only has about 3-8 gold per turn available for trade in that era, so I will get a lot less money for my resource if I sell it. It's a bit of a trade off. I know that the AI can grow larger cities because of my resources, but I get some gold. Will I do the deal or not? Maybe the AI shouldn't do the deal if it really has enough health and happiness resources. Probably that logic is missing from the AI. But know that I usually would buy a health or happiness resource for a meager 5 gold per turn which is what I typically get from the AI.


I don't know if the strategic resources are valued right. I once had a vanilla civ game where I had to pay something like 300-400 gold per turn for a resource like iron. At that time that was half my economy (the total gold income was something like 700 gold per turn at 100% tax which was also partly required for maintaining stuff). Another way to get such a resource is by trading 3-4 happiness or health resources for it.
At the same time I often see two AI civilizations trade 1 iron for 1 cow or something like that. So they do these trades with oneanother but not with the human player.

So, if anything, the human player is penalized in resource trading. I'm not principally against this penalty since the human is far smarter in getting the money from the AI. But I also don't see a reason to let the human player get even less gold for their resources. I don't think I'm willing to part with my cow and help the AI grow bigger cities for a meager 1 gold per turn.

So, you should not only look at the human player advantage of these trades, but also at the AI advantage (a maximum could work, but I've never seen the AI pay an amount which I wouldn't pay myself).
 
A big thanks to Blake and the betterAI team who improved the game so much !

I just finished my first game with betterAI. I won on Monarch, standard size, continents, 11 opponents, permanent alliances on, other options off.

I had few troubles winning but the random leader I got was Caesar and so I made my first Pretorian Rush and it helped a lot to vassalize / destroy everybody. The build was the 12/07 one (or was it 12/09 ?). The game was very pleasant.

The only real complain I have is the AI didn't seem to defend its cities with enough units. I know it's very hard to stop the pretorians, but here there were generally too few defenders (2 or 3 on monarch is clearly not enough)

To make a quick recap of my first impressions (I need to play again to be sure) :

*) Main Positives :

- More aggression AI vs AI than in 2.08. I'd like even more inter AI alliances and vassalization but it's ok. An AI destroyed another one on the other continent I didn't know (I got a message), and Brennus vassalized Monty there too.
- Much better use of the ressources
- Good places are chosen for the AI cities
- AI is much better in science
- AI seems to whip and draft much more, which is a very good thing

*) Main Negatives :

- Too few defenders in the AI cities (generally between 2 and 3)
- Maybe not enough emphasis on production from AI compared to science (not sure, but I was very often #1 in prod)
- Some strange civic choices were made for a long time (hereditary rule, caste system...) but maybe there was a good reason I don't know (I doubt...)
- I captured several workers in war time, maybe there's a bug somewhere on this part
- AI Stacks are sometimes easily destroyed in the battlefield
- I chose Emancipation but then some AI civs didn't discover Democracy before a long long time (Carthage), so I guess they were heavily penalized by my Emancipation happiness penalty... (not sure they were, didn't use the world builder to see, sorry)

*) The things I would like to see :

- AI should probably try more to make pacts, peaceful vassals or alliances, with other AIs or with human. I was much stronger, but had they allied they could have resisted better.
- AI should really aim for victory and react when they see another one is winning... If we see an AI winning, we surely try everything we can to at least delay the victory, so the AI should do the same, whoever is winning (nukes, massive invasion, etc.)
- Big naval invasions
- AI who really aims for diplomatic victory
- AI who really aims for conquest / domination victory (I know, they would need a plan and it's terribly hard to code, but these are my dreams :D )

I will make another monarch game with the latest build. I will probably have much more trouble because my random leader won't be from Rome this time ;)

Anyway, keep up the good work, the game is really great now !
 
So please Iustus help me,

Altering the warmongerrespect values will screw up AI decision making or not?

Thank You!!!
 
By Blake,
If I can think of a really good all-encompassing way for valuing gold then something might happen, but for now I feel that simply being stingy with their gold is closer to "good balance" that the 1.61 state which was definitely unbalanced.

Maybe I missed the point of the Mutineer, but I know I'm stingy with my gold. If the goal of the AI mod is to have it try to mimick how a human adversary would react in similar situations, then stingy is definitely one of the places to start. So the question really becomes what do humans do with their gold inside the realm of CivIV. We use it to upgrade units in emergency situations, we use it to run deficit spending on research, we use it to rush buildings and units if we deem it necessary (after US of course, mid to late game w/o Pyramids), and we try to trade it for techs or resources that will work to our benefit. I believe this is exactly what Blake has tried to do if I have been reading correctly the 875 some odd posts that I have tried to wade through. Can further improvements be made(?), possibly, and I'm sure they are being worked on as we speak. I know absolutely nothing about programming, but I have been playing Civ in 1 form or another for about 12 years, and the first time I tried Blakes changes (see my post above) I got wacked on a difficulty level at which I am routinely victorious in almost any way I choose to win. Coming from the point of view as a non-programmer, the changes do exactly what they are supposed to do, make the AI a more difficult adversary. If I could offer anything that Blake and his team had not considered, it would just be a way to prioritize the way an AI values the gold, but I'm pretty sure he has already thought of that.
 
Back
Top Bottom