AI city placement and misc. suggestions

[civ4-advciv-oracle-bug]

This example is more of a question, it is example 29.
After capturing Ragnar's capital, i notice he has no granary (nor barracks) in his capital (no building at all in fact). (His other 2 cities that i destroyed were also empty if i remember correctly).
Shouldn't the granary be given a higher priority? Especially if he is a warmonger, he should want something like a granary and maybe also slaving barracks.

If he is a warmonger, he would want promoted units.
I understand maybe for the early archer rush, but except for that in 100+ turns still not having any building in his capital that has tons of hammer + sheep is strange.
Would you consider making granary higher priority or/and barracks for warmongers at least.
I also noticed this in another save file where captured capital (aksum if i remember correctly, i think it was example 16 btw i mean if i am not mistaken i mean anyways), i could provide more save files if you want to investigate.

If this is known or/and expected, i would suggested getting either a granary or barracks or both, especially for warmongers, if city is a productive one.
I assume he slaved a few times to produce all these swordsmen, so it would be inefficient to produce them manually while being saturated in food growth i mean.
Would you want to make AIs have a higher early focus on either of these buildings, notably if they are warmongers, at least in their production cities (instead of producing units drop by drop and not slaving, or slaving and not having a granary to maximize it for example i mean anyways)?

Perhaps this would incorporate well with the more long term war declaration instead of rushing it then withdrawing and having lost more in the long run than building a few more units and maybe one fo these buildings or both buildings first? Perhaps a more defensive early game but that aims at conquest later, which is not the same as being passive or too friendly i think i mean. Also this would be adjusted by difficulty so maybe it would be a function of already produced units and production rate or something i dont know how you would code or implement it i mean these are suggestions i mean, but i think one way or another getting the granary, or rather skipping the granary or/and barracks is bad for a warmonger in particular, after the first early invade fails he would have 2 less buildings than me who survived, so better invest a bit more and have a stronger offense and stronger production potential some more time later?

What do you think of this? Thanks,

 

[civ4-advciv-oracle-bug]

Here i mean is a weird behaviour of AI settler and garrison which i think would be nice if could be improved, i documented a few save files of what i think is edge cases (meaning the action happens soon after this save file by just pressing end turn one or a few times + is a quite typical action of this behaviour), i hope they can improve AdvCiv i mean.
This is example 30, i added a few save files and a screenshot for the city spot.

Here are the main points i want to discuss and hopefully so that they would be improved in advciv i mean:
-b.1.1) the city plant spot, in the screenshot of example 30, many cities are really nicely planted, not planted on metal, and trying to maximize food or hammer yield while planting on weak tiles (such as tundra or plains i mean it seems anyways) as much as possible. The cities i have destroyed before were i think not optimal and i have provided some of them in previous examples. Here i think just planting izmir one tile south would be much better, as the 2nd ring of izmir mostly has desert or low food, there are many hills for production but not enough food to exploit these so they become useless i think. So, i mean i would suggest the spot 1 in screenshot, which is where i planted my city (after barbarians destroyed theirs but it's a point after this one). In spot 1 many grassland tiles can be turned into farms, also less pressure so i think it's a better spot, even after civil service would be discovered. Use save file example 30.b.1 to investigate its decision to plant here.
-b.1.2) it would be interesting to investigate why AI decides to plant here, but in the a.1 point in same position AI runs away ("shy of my army" i mean?) I think running away is the correct decision due to a barbarian swordsman being nearby, i am not sure but i think they should see him due to hill vision, if they dont see him, still avoiding to plant may be better, in advciv a bit too much cities are destroyed, but Suleiman AI has a strong army or so it seems, so i would also suggest to make AI (if possible i mean anyways) have a bigger portion of its army/garrison in edge cities, maybe a minimum of say 6 strength or +2 relative strength (with bonuses and maluses) as compared to a barbarian invader axeman of 5 strength (maybe i mean its a suggestion/example i make i mean anyways) to be safe on edge cities would enough to repell barbarian invasions? I ask this in the point a.1) below.

-a.1.1) here as said just before (in save file example 30.a.1 now) AI chooses in same position to not plant, seemingly only because i put my army here. I think it would be interesting and maybe hopefully helpful to see why AI ran away and did not plant izmir (as i said i think it's the correct decision, instead of having the city destroyed just after plant). Find and i mean enforce the mechanic that made AI reluctant to plant here or at least now, in a way that doesnt make AI expand too slow, ideally find the balance between expanding enough and not having too many cities be destroyed soon after plant due to low garrison. This also happened in many of my save files and makes the AI quite weaker. Maybe it would incorporate well with the patient AI strategy of not rushing someone too soon, but instead build more units, and ideally have a repartition (if possible to code i mean) of these units so that more are on the edge cities maybe.
-a.1.2 and a.2) unrelated, but in same save file, investigate why AI does this weird settler back and forth (same as in example 28), he comes then runs away in 375 AD in save file 30.a.1, then does the same thing and settler comes again and runs away in save file 30.a.2 in 475 AD. I think this is really inefficient, maybe safe to not have city die, but i feel/think it's something else (contradicting orders/logics?) that make the AI both decide to come plant, and then decide to run away not plant just after. Is it related to how AI declares war then retreats immediately after without attacking? Maybe there's a global way to fix or workaround or approach i mean this issue if wished/possible i mean for advciv?

What do you think of these and do you have ideas on how to fix them? I think it would make the AI quite stronger just by enforcing these few rules, if i sum up this post :
- focus more on the grassland farms especially if there are many hills for city plant, not the desert/plain
- weird settler "hit and run" and running away / "shy of my army"?, inefficient, either plant or not come
- planting with too weak garrison, i think Suleiman AI has enough military but it's in other cities, plus he is not at war with anyone unless i'm mistaken, so mobilize/allocate a higher percentage of garrison in edge cities (less in center cities), for example if there is no rival in this direction (only barbarians) then add only a bit of garrison, if there is no barbarian add a bit less than that, if there is a nearby rival or more in this direction of edge cities allocate a bigger part of the army, i don't know if it's feasable to code and possible and/or ii mean wished for advciv but would be nice if possible
- resulting of this, avoid early city plants that are then destroyed, at least a bit more, in a way that doesnt slow down AI expansion due to being too safe, i think focusing on available garrison and making AI have a more patient early game may allow that, so that he strikes with a decisive attack a bit more in the middle game after developing safely, something like this i mean.

What do you think of these? Thanks,

edit: maybe the settler choosing to come settle then running away being "shy of my army" or so it seems is also related to early war declarations then the stack being shy of my city garrison then running away, then coming back with a bigger stack in same war much later then soemtimes rerunning away being "shy of my army" i mean anyways? If it is related, is there a more global approach to it maybe i mean anwyays?

edit 2: a workaround that buffs specifically more the AI than the human player (without affecting human vs AI) may be to give a small bonus to archers specific against barbarians, maybe something like "+25% city defense against barbarians". Since AI has more archers than the human player on average, and human player is less likely to lose cities to barbarians (typically its a lone swordsman/axeman barbarian vs archer that quite luckily kills a solo archer guarding the AI city (typically new cities too so often raised i mean anyways)) that , screwing the AI, so maybe just a small buff to archer city defense against barbarians may be a workaround to possibly i mean mitigate maybe the effect of AI losing many cities to barbarians early which hurts it, without affecting human vs AI later warfare if there is any. I'm not sure it's the best approach but it would be kind of a workaround. It's not perfect, maybe not ideal, maybe there is a better and more global way to address it, but maybe as a workaround it may help the AI not lose too many cities early? What do you think of this i mean? I think it's cool that the barbarians are so strong early and would be nice if it can stay like this and threaten the human player too, just buff the AI a bit by its behaviour (more aware/more garrison maybe) or workaround (buff slightly (solo) archer city defenders (mostly helps AI as it has more archers i think in general than humans players but i don't know for sure). It is not ideal though that it helps the human player however, but a human player would be able to workaround barbarians more easily i think so less affected. Thanks,

 

[civ4-advciv-oracle-bug]

This one seems more like a bug but i don't know if it is maybe intended?
It is example 31, i provided a save file and screenshots

When trading with Charlemagne, if i ask what he can offer me for cows he says nothing, but if i manually propose from him 3 gold per turn against from me cows then he says it's his final offer (meaning he is ok i mean anyways). I wanted to trade cow for 3 gold per turn as shown in the screen shot and foreign advisor/ressources, but i was surprised he says it can't be done if i ask him what in exchange for cow, but if i input manually 3 gold per turn then he is ok strangely?

Is it a bug?
Thanks,

edit: added a few more screenshots for exhaustiveness i mean anyways

edit 2: a similar issue was encountered in example 68.2.a and 68.2.b that i added too (with screenshot), see example 68's post for details

edit 3: there are other trading bugs, for now they are in total: examples 31, 34, 68, 105, 106, see their respective posts for details, thanks,

 

[civ4-advciv-oracle-bug]

This one i mean is quite silly of the AI and would be nice if could be improved
It is example 32

After ending the turn, AI will attack my swordsman with 2 siege units. I think generally solo attacks by one or a few siege units (without their stack accompanying them or without attacking a big enemy stack or without setting up a city attack later) are not profitable, especially since here he wastes a catapult and a trebuchet to just attack one swordsman and one wonded chariot.

If possible/wished for advciv, i think i mean the benefit is significant enough to have a sort of more strongly enforced rule i mean that siege units should not get "baited" and i mean attack (especially if not in context of city attack or weakening a big enemy stack with splash damage i mean anyways) enemy units solo, especially if it's just a few enemy units, better retreat and wait for a better opportunity or defend.

He wasted 2 siege units which gives me quite an advantage, also he should know based on me just capturing his city just the turn before i mean anyways that i have a big stack nearby so even if he kills this swordsman somehow or/and chariot he would lose 2 siege units immediately after, perhaps even baiting me or giving me tempo to attack him soon after, so i think if possible/wished for advciv more strongly enforcing siege units to not attack on their own unless it's for a city attack or they are with a stack nearby (but it doesn't seem to be the case in this save file as i just killed a few of his units so i don't think he has many more moblizable immediately) or even if they are solo as long as they damage a big enemy stack (maybe 5+ enemy units or in context of invasion maybe i mean) it may be worth it, else keep these siege units with a more defensive or supporting i mean behaviour may give more benefits for the AI and make it stronger i mean rather than solo suicide attacks to attempt to kill isolated enemy units, thus losing the siege units for no real benefit i think i mean anyways

What do you think of this? Thanks,

 

[civ4-advciv-oracle-bug]

I have another example of weird back and forth with AI stack, but this time he did not attack.
(edit3: after withdrawing he comes again with his stack close to border in 1060 AD during his turn so maybe he'll declare war in 1080 AD or soon i mean anyways, still weird back and forth)
(edit4: he then rewithdraws in 1080 AD without declaring war, very very weird i mean, i assume it would go like this for a few turns or maybe he gives up i mean or maybe finally declares war i mean anyways, would be nice if this behaviour or/and logic or/and his strategy i mean anyways could be improved as said in the post i mean anyways)

It is example 33

In save file example 33.0, after ending turn Japan AI moves his stack at my border so i was expecting him to declare war next turn. Strangely or coincidentally, at this turn now i am no longer his worst enemy, and next turn he withdraws his stack, in save file example 33.1 (after ending turn).

I don't know if he withdrew due to me no longer being worst enemy, or if he saw my incoming stack that i had prepared i mean (does he have a spy in my territory i don't know or maybe the hills view was enough to show him more enemy units than he expected so he withdrew?), but this behaviour seems quite suicidal. Regardless or if it's whether i'm no longer his worst enemy or if he thinks he is stronger (relative to me) as he actually is, he almost declared war which would have been bad for him, in some other contexts i could wipe his troops and counter attack. I think it's better to play it peaceful until he is sure he can win a military war (and prepare and bump i mean the troops production during this time to have more troops than me then finally i mean launch a big attack right at once).

edit: but since this happened in other maps i played, and there was no diplomacy issue to make his renounce war declaration if i remember correctly, i would assume there more factors at play i mean anyways than just diplomacy (no longer worst enemy i mean anyways) that made him renounce war declaration and alternate between these. It is still inefficient though and would be nice if could be improved i mean anyways.

As i said in many other similar examples, this is disadvantegeous to him also because now the surprise effect is lost (even though i was prepared and expected/thought he would declare war on me maybe) due to past war(s) i mean and his personality i mean and etc i mean anyways.

So would you consider making the AI more decisive in its war declarations, instead of doing "half assed" i mean for lack of a better word i mean wars i mean, that become a flop and disadvantage him later in many ways i mean anyways, if only he built his stack more and only declared war when he was fully ready and determined to make profit of this.

This example is also interesting because of the no longer worst enemy, i don't know if he changes his mind because of this, maybe in the debug or log or some other things you can see what influences him to renounce, and in save file 33.0 what made him decide. Maybe making him angry would show if it's the diplomacy (no longer worst enemy) that made him change his mind (for his own good i mean anyways). But i think making the AI more patient for greater gains and decisiveness/efficiency may help make it stronger i mean. Also, to investigate what seems like contradicting logic (but i am not sure it's just my impression i mean anyways) of back and forth in many situations and types of units, making it inefficient and weaker AI (he loses some advantages of tempo as comapred i mean to if he did thigns directly and decisively i mean anyways). Would you consider to improve these behaviours, investigate too i mean anyways, and possibly changing a bit its strategy for more patience and long term aggressiveness (delayed) but later, if it's possible/wished for advciv i mean? Thanks.

edit 2: even if it's samurais and i don't have macemen yet, 8 units is a bit too little i think so maybe he overestimatde himself? Just a few longbowmen (which i already can upgrade from archers i mean anyways) would make his invasion most likely to fail i think i mean anyways. Would you consider make him require more strength differential or/and better estimate enemy potential for defense?

 

[civ4-advciv-oracle-bug]

I have another example that is quite weird related to trading i mean, it is example 34
I have provided a save file and 2 screenshots for comparison i mean anyways in same download folder

If i give Charlemagne a few trading items from me against 30 gold per turn from him, he is ok.
But if i give him the same trading items and i add on top of it fish for free (meaning it would cost him the same price of 30 gold per turn for the same items + one more item added for free), then he is suddenly not ok.

This is quite weird, is it intended?
It costs him nothing to take another fish from me so why would he refuse it, it's not realistic as a real player would take anything it can gladly i mean anyways.

Maybe it is because they don't want to take too much items in trades? For example i noticed that if an AI wants a few items and i trade one or 2, then he doesn't want the others anymore.
Maybe it is to nerf abusive or too much trading from humans against AIs?
However, in this specific case, taking fish costs him nothing, so he has no reason to refuse it. Maybe if i ask more gold of him he may think it's too much, but for the same price he should accept as it benefits him to get one more item for free. Also, maybe after i make a trade with him it would be expected/understandable i mean anyways that he doesn't want to trade (too much i mean anyways), but since i did not trade yet and it costs him nothing, i think this is not optimal, he should just accept as much as he can from me i mean even if it's not fair for me i mean.

Would you consider fixing this, or is it maybe intended behaviour of the AI i mean (but really weird as in not benefitting him i mean)? Thanks,

edit: maybe you could name this fixed behaviour for example i mean "make AI not mind being more greedy if it does not cost him anything more" if you would want to fix it i mean anyways.

edit 2: a similar issue was encountered in example 68.5 that i added too (with screenshot), see example 68's post for details

edit 3: there are other trading bugs, for now they are in total: examples 31, 34, 68, 105, 106, see their respective posts for details, thanks,

 

[civ4-advciv-oracle-bug]

I have another example that i think would be nice to improve
It is example 35

Here if you press end turn Japan AI i mean gets baited and goes to defend/capture back a weak city, abandonning i mean while doing so his capital which is of much higher value (capital has a few wonders and big population, the other city only a few buildings at most if i remember correctly).

To take his capital after you end turn i mean, batch upgrade all your cataphracts i mean to max strength promotion, then select all stack and batch attack without changing the order, then you take his capital which i think is very very unoptimal of him.

Correct behaviour would be of him to focus most of his garrison on strongest cities, here the capital has much higher value, he could probably defend it with all his stack, and the other city was lost anyways i mean so why get baited to only lose your capital that was nicely guarded and much more important i mean anyways?

Would you consider fixing this too in advciv i mean it would be nice.
Just make AI focus on its strongest cities, and not get baited by a few units or a few weak cities, especially if by doing so he splits his troops and becomes weaker (on top of losing key units/cities later or weakening his advantage i mean).

Thanks,

edit: i have provided example file 35.2 with the result of doing these actions: his capital i mean city is captured. Would be nice if AI could be improved to not get baited by weak value units/cities and focus on stronger ones (especially if he is weaker in this example i mean anyways) to get more value of his actions i mean anyways

edit2: it is also very weird that at turn 200+ his capital still doesn't have granary nor barracks yet, i used a spy in another save file and saw that he produces many samurais and military units, why did he not spend just a few more turns to invest on barracks and possibly granary too.
I really think these buildings should be given higher priority especially in later game when they cost as much as one unit or even less for an AI that has 15-20+ units already, he could have just produced the barracks first and get nice promotions i mean with it, granary would provide earlier growth, it would involve a bit more patient early game for a better late game, same as in example 29

 

[civ4-advciv-oracle-bug]

I have another example very similar to example 32 and example 35
It is example 36

Here a solo pikeman gets baited and chooses to attack my stack, making him so that he will die next turn.
The pikeman would have been so much more better as a city defender with the rest of the garrison (in this example i may ultimately still capture the city, but as a general rule to have units not get baited but stick with their stacks would make AI stronger or keep its advantage or not become weaker than it is i mean anyways).

It is very similar to example 32 where siege units get solo baited very inefficiently and would have been so much better as supporting units with their stacks instead of these solo/suiciding attacks that only harms the AI and makes me stronger i mean anyways.
In example 35 too AI gets baited by a weak city and abandons the defense of the much more valuable capital city, despite seeing my stack incoming on the capital.

i think it would make the AI much stronger to understand these not get baited things, at least behave in a way that does not cause suicide of solo units for no purpose with the only result of making the opponent stronger, either attack with a stack, or 1v1 if odds are good or benefit is good, else do not solo attack into enemy stacks or a unit near enemy stacks.
Would you consider to improve this in advciv, i think it would be very nice if it could be improved, but it is my suggestion i mean so i don't know if it would be possible/wished for advciv i mean anyways.
Thanks,

 

[civ4-advciv-oracle-bug]

This example 37 is even more critical.
It is similar to example 35.
This example 37 is what happens the turn just after example 36.

In example 37 i mean, after killing the pikeman in example 36 with one unit, moving my stack forward to his capital but in another tile than where this unit i used to kill the pikeman is, and then ending turn, then (starting position of this save file) if you end turn AI makes the critical mistake to get baited by one wounded units and waste 14 units to kill it, instead of defending his capital. Now his forces are split and i am in attacking range of his capital that only has 8 units. What's even worse is that the units are in a not roaded tile i mean so i can move my catapults leisurely if i want and he would need 2 turns however to comeback defend his capital. But the general rule of abandoning capital for one unit is very critical i mean.
He could have had 22 (8 + these 14 that he didn't move to capital but used to chase one unit) capital city defenders (+100% city defenses), but instead i am free to capture easily his capital now or/and kill the 14 units separately that are now sitting ducks xd i mean as one of you said this word i mean anyways. It's a big blunder i think, also not realistic, as in real life no general or commander i mean would chose this silly splitting decision that kills the entire army or country i mean xd.

So it would be nice if it could be fixed that AI does not get baited by one or a few units or one weak city and instead attacks or defends more based on profit or value of cities or units i mean.
Thanks,

edit: although not as critical, i observed a similar behaviour in example 67 of a big stack leaving cities and that can be "sniped" from outside cities, while other cities are quite defenseless and threatened to be attacked. It may serve as an extra data point to observe or test, else it may be fine to ignore this example 67 i would say i mean anyways, thanks,

edit 2: actually the example 67 is not far from as bad as example 37, i added save files 67.2 and 67.3 of what happens in the next turns and detailed this in the example 67's post, ideally he should avoid pushing too far when he is too weak and on threat of attack/invasion, and instead focus on defense or/and ask to be someone's vassal to make the game still competitive i mean anyways maybe, thanks,

 

[civ4-advciv-oracle-bug]

This example 38 is also very similar to example 36.

When you end turn, lone AI pikeman gets baited by my incoming stack instead of being much more effective as a city defender. He could move to defend the capital city or stay in his city, but considering i have a nearby stack of cataphracts threatening his capital, it's suicidal of this AI pikeman to choose instead to solo attack my stack instead of defending cities (here defending his capital would be the best target i would think i mean anyways).

In this specific example i would still win and capture his capital, but as a general rule to not get baited by one or a few units when the core focus should be to defend. In similar positions, it may make a big or critical difference in war outcomes (possible stalemate instead of dying, or keeping advantage if AI is already stronger) if AI better handles these kind of war strategies i mean if possible even in a not exhaustive or advanced i mean way of handling it would still make a difference i think i mean anyways.
So it would be nice if this behaviour, similar to the others, could be fixed or improved i mean, but it is also just my suggestion so i don't know if it would be wished/possible for advciv i mean.
Thanks,

edit: my mistake, pikeman attacked into my cataphract stack not into my catapult stack i mean anyways, but the principle still remains i mean that he would much more ffective as a city defender instead, in the capital city, even if it would not change the outcome in this save file, as a general rule it may help a lot to have the AI have this kind of behaviour i mean anyways, if wished/possible for advciv i mean anyways. Thanks,

edit2: example 43 is very similar to this, see example 43 if need or want to test it too i mean anyways, would be nice if AI could be improved to not gradually or totally lose its advantage or worsen its bad position by these units getting baited for no profit but only loss i mean anyways. Thanks,

 

[civ4-advciv-oracle-bug]

Also this is a small UI request i mean but could you make great people production visible in buildings and wonders thumbnails i mean

i attached example 39 files screenshots for the difference with and without (quickly edited so just to give a general view). You could use save file example 38 and zoom in constantinople (my capital city i mean anyways) to see the city default view i mean anyways.
It would make it much much easier than having to remember every time which wonder does what, so a lot less strain i mean if possible/wished for advciv i mean anyways.
Thanks,

 

[civ4-advciv-oracle-bug]

Also, this is not an example, but just to say that except from the few examples i pointed in 24, 26, 30.a.1, it seems that the AI city plants are all (or most i mean i would say anyways) very very nice. Since i have the world map for quite some time i mean anyways, for example in example 38, i don't have much to say about how city plants could be improved, they have all a nice balance of food and production trying to maximize ressources, and no one is planted on metals but always near them.
So for the most part i would say it is very satisfying and i am thankful to you for that i mean anyways, i don't know if it's the result of the changes recent or not i mean anyways in advciv 1.12 or if it's just many coincidences, but it is seems the city plants in advciv 1.12 are really nice overall i wanted to say that and thanks to you. I pointed a few examples of where it could be improved i think i mean.
But for the most part if i don't say much it would mean i am (very) satisfied about the rest so thanks for that i mean.
I will continue to add examples of what i think could be improved i mean as long as i like and want it i mean anyways, but it was just to say that, so i continue playing from now if i want/like i mean anyways.

 

[civ4-advciv-oracle-bug]

I also added an edit 2 about example 35 (https://forums.civfanatics.com/threads/advanced-civ.614217/post-16769898)
It is weird that Japan AI still has no barracks nor granary in his capital city at turn 200+, i think he would gain much more benefits having barracks or/and granary, see the edit2 in the linked post for more detail i mean anyways, same as in example 29, thanks.

 

[civ4-advciv-oracle-bug]

Sure i understand, actually i'd rather that you focus on implementing the fixes rather than reply in detail to me, because it would take a lot lot of time just to reply to mine, so if possible i don't mind that you skip replying to me in favour of reviewing these examples and code.
As for the examples or posts of mine i mean that you would not focus on, all i can do is post examples that i find significant and let you choose i mean anyways, so thanks for that.

I played another map in advciv 1.12 and it seemed somehow better city placement though, but yes it may be coincidences or slightly/quite better, anyways thanks regardless of how deep you go into it i mean, but as i said if i would say what i would prefer is for actual code to change rather than long replies to me i mean but of course it is my suggestion only i mean anyways so in all cases thanks

 

[civ4-advciv-oracle-bug]

Also about city placement now that i have captured Japan AI's cities, i noticed he spaces some of his cities a bit too much and could have squeeze 1-2 more cities in his territory i think
I provided example 40: 2 screenshots and 2 save files.

From what i see, early he rushed to barrage Boudica AI at his northwest side (tokyo in 1600 BC save file 40.1 turn before he plants it) and me at his south east (satsuma 800 BC 1 turn before he plants it save file 40.2 provided)

However about 40.2, although the gold is nice, he could have went one tile north, gaining access to coast and about the same yields (except no gold ressource early). If he did that, he could have the desert hill iron mined (but not that good though just that he would have good enough production even if going north i mean anyways), and doing so now there is more room for one or 2 more cities, one of which has gold in it. Considering how tight the map is, he would have greatly benefited in the long run to plant one more city i think south east of tokyo. I understand that his economy may be tight too i mean early anyways but at turn 200+ he still did not plant neither of the city locations i proposed in screenshot 40.2, which i think would have benefitted him especially if he played a bit more patient early game.

About 40.2 and the same question for screenshot 40.1 too, the east location i propose for city plant (similar to example 2 that i sent a long ago) is i think really good due to abundance of food and new resources he doesn't have, of course not early but sometime in turn 100+ he should have enough gold to afford a city there, assuming he does a moai statutes i think a city there would help him too (he could slave a few citizens to get moai statues sooner and pay gold per turn for stones if not too expansive? Another advantage is he would gain access to clam and cow (health for all cities) that he doesn't have unless i'm mistaken or he could export them if he wants so all in all whether for health or/and gold export he would benefit from these i think

So i provided the corresponding save files so that you can see how you could influence the AI's thinking so that it :
- spread less when land is tight and more cities are plantable if they are good enough, especially if they bring advantages (don't mind the small quite bigger early investment)
- plant more cities when they are good enough and there are enough good enough workable tiles, it is related to previous point : concentrate more cities closer to each other (without too much overlapping if have too, here too many tiles are wasted i think which would have helped japan AI be more competitive).

Would you consider improving these too? What do you think of these?
As i said in previous message i'd rather that you focus on reviewing these examples or/and doing code related changes if you deem/want them necessary/worth/wished i mean anyways rather than giving extensive replies to me that would spend all the enrgy xd, in fact you can skip replying to me at all as what i want most is to play against the AI xd so as long as there is code all good for me, of course if you have the extra energy i mean to reply to me all good, but if i had to choose i mean i would prefer no reply to me and most energy on code and reviewing xd, so in all cases thanks i mean, thanks,

note: in the download folder there are .7z archives that you would need to extract too else you may not notice ingame the save files in these archives i mean anyways, thanks,

edit: similar to example 2 spot 2, not to example 5 was my mistake
edit2: my mistake he didn't have cows too (same as clam) unless i'm mistaken, so he could benefit from the east city of cow clam even more than i thought.

 

[civ4-advciv-oracle-bug]

I found another example that is quite critical too. It looks like a bug unless i am mistaken i mean anyways.
It is example 41.

Available ressources are not improved at turn 200+, but instead have a weird fort on them.
I think they belonged to Japan AI in osaka before i captured it i mean anyways, or maybe the gold was to Boudica, but regardless of which AI it is, the ressources should have been upgraded a long time ago, gems could have given max +2 happiness or export it if extra, same for gold, but they wasted time building a fort instead as if they confused it with the mine upgrade
You can also look at example 40.3, the cow is not in Japan AI's territory however it has a fort in it (but no pasture since i assume there was no city there). Why build a fort if no one is gonna invade there (did they confuse it with a pasture?)
Also, there was no war between Boudica AI and Japan AI i mean anyways so i don't think it was pillaged (else they would have repaired it long ago, i noticed when i pillage their iron mine for example (that had a fort) they rebuilt the mine immediately after war).
The issue may be to prioritize forts way too highly (why not build them if everything else is done but not before unless cirtical strategic spot but i doubt i mean anyways) or simply confusing forts with ressource upgrades, or altogether forgetting/not being aware of ressources to upgrade (but it seems more like a confusion here i mean anyways).

By the number of turns throughout all the game, i think AI would be a lot stronger if it just prioritized upgrading ressources first or as soon as possible, but even if not immediately to not confuse forts with ressources upgrades which seems to be the case here i mean. Build fort if you want but first upgrade the ressource i mean for the AI anyways.
To investigate this you can any save file between examples 24 and 41 as they are all about the same map so you'd have their view at different times of history.

Would you consider improving/fixing this?
I think it's quite a lost opportunity for the AI and not intended unless i'm mistaken, it may make the AI quite stronger throughout the game i think i mean anyways.
Thanks,

edit: additionally to fixing this if you want/wish/can fix it i mean anyways, you may want to give forts a much lower priority, i assume it would be quite rare that they play a critical role, it may be a workaround to this issue (but ressource still needs to be improved), while also improving worker efficiency, as they spend more time boosting their cities or econoy rather than buidling pointless forts (which i noitced they do many times, it may be fine if they have nothing else to do of importance i mean anyways in any city nor in their territory at all, else it makes the AI quite much weaker)

edit 2: i added example 81 that shows the same problem but before the fort is build, and the ressource on this tile was already improved (sheep with an existing pasture), if need to refer to it it may help maybe i mean anyways.

edit 3: linked to example 85 where an AI worker builds a cottage on unimproved banana instead of plantation or nothing, see example 85 for details, thanks,

 

[civ4-advciv-oracle-bug]

edit2: it seems that i got the modifier of "military struggle" in 1740 AD (maybe had it before did not check) so this example may not apply anymore, however it may have been nice to get it sooner after war declaration just like how war declaration immediately modifies relationship with friends of the declared war on i mean anyways. Also, the suggestion about number of cities increasing this relationship modifier is something i would still like to ask as explained below but again i mean it is a suggestion that may or may not be wished/possible for advciv i mean, thanks,

I have a suggestion about relationship modiiers which would be really nice and hopefully not too hard if wished/wanted for advciv i mean anyways
It is example 42

Here Charlemagne AI is at war against 3 civs and losing cities every few turns.
I am at peace with him currently.

In this save file, i am about to declare war to Boudica AI.
Boudica AI is one of the civs who is at war against Charlemagne (they declared war on him btw i mean anyways).

What i would like is that from this save file, declaring war to one of Charlemagne AI's ennemies would give me a +1 relationship modifier with Charlemagne AI, i think it would be very nice too i mean anyways:
"+1: The enemy of my enemy is my ally".
It would be very realistic too, as in real life surely alleviating the pressure off from one of his rivals/war enemmies would surely help him and advantage him/make him happy regardless of my motives.
It is also quite similar to the "Our military struggle brings us together", but would not require the extra conditions that seem necessary for me to have it (i don't have it after declaring war here).

I'm thinking it should be temporary and only for the duration of the war, but maybe it could fade slower after war too? (i think rather temporary only during war, if you want to add it i mean anyways).

Also, it would be even nicer if i would get a +1 per city i capture, for example if i capture one city from Boudica AI it would be:
"+2: The enemy of my enemy is my ally (+1 for cities captured)".
And if i capture 2 cities it would be:
"+3: The enemy of my enemy is my ally (+2 for cities captured)".

To prevent from abusing this mechanic, the +1 modifier can only be gained once per city, so i cant abandon a city and capture it just to raise artifically the modifier.
Also, losing the city would lose the +1 for city captured, i think it's quite realistic too, as he would not be as enthusiastic about me if i failed to beat his enemy (but still quite happy i am at war against his rival).

What do you think of this? I think it would be very nice to add in advciv if wished/possible can i mean anyways.
Thanks,

edit: in parallel to that, maybe you'd want to make the war penalties a bit harsher, for example a global "-1: You captured cities from our friend" that would be realistic too i mean. However doing it -1 per city may affect relationships too much, so maybe this would be temporary only for the duration of the war and then it might be fine (and realistic too i mean, as if they watch their friend dying they may be more compelled to intervene, it may also be good to trigger more opportunism of the AI from the exhausted warmonger but not too promote this effect too excessively if warmonger ends war sooner). What do you think of this i mean anyways? Thanks,

 

[civ4-advciv-oracle-bug]

About relationship modifiers, i would also like and i think it would be realistic if when an AI declares war on someone (whether it's another AI or a human player), the AI that declared war would get -1 modifier "You declared war on our friend" from the AI that got declared war on's friend(s).

You could take example 25 or 19 save files or similar example save files where AI is about to declare war and try to implement it if wished/possible for adviv i mean anyways.
I think it would be really nice and also realistic though, as unless i'm mistaken only the human player has such penalties which doesn't seem sensical i mean anyways nor fair.
Thanks,

 

[civ4-advciv-oracle-bug]

I also added example 43 very similar to example 38.
Near tolosa, a lone pikeman gets baited instead of being much better as a city defender instead, which i think i would be nice if AI was improved this way to not get baited and does these solo or group suicide attacks especially when defense or sticking with its stack is better
I will also add an edit in corresponding example 38 post.
Thanks,

edit: unrelated to this example, i added example 44 as a screenshot.
Is it normal i mean that on example 44 screenshots, i cannot build a watermill on this tile, but there is a river nearby.
Doesn't the fresh water come from the river?
There doesn't seem to be a lake nor oasis or other source of fresh water that makes this tundra have fresh water (plus the tile has one more gold so i assume it is linked to the river right?)
I may be mistaken though but if i understand it correctly is it a bug? If so, i would like to be able to do a watermill if i want i mean, but if i am mistaken as said in previous messages i'd rather that you focus on reviewing the examples instead i mean even though i don't mind or would like a reply but considering all the examples to review i really rather you'd focus on the code and review i mean but it is only my suggestion i mean anyways, in all cases thanks,

 

[civ4-advciv-oracle-bug]

I think i found a bug too but i am not 100% sure.
It is example 45.

A monument gives 2 culture per turn for an unknown reason.
In 1400 AD monument in city 2 i mean anyways only gives 1 culture per turn.

Then if you skip turn twice in 1350 AD the monument culture becomes +2 per turn.
Is it because my archer won against axeman??
I tried slaving the granary sooner, cancelling chopping, i have no idea why monument gained one extra culture.
Is it intended or is it a bug.
It's also not visual as the turn after 1350 AD city 2 actually gained +2 culture going from 25 culture to 27 culture if i am not mistaken i mean anyways.
I attached save file and screenshots.
Thanks,

edit: split this into 2 examples: save file example 45.1 (in 1400AD) when monument still has 1 culture per turn (unrelated i also removed city plant marks with the alt+X shortcut that you told me thanks for that too i mean anyways but i mean i confirmed the issue still happens in this save file by skipping turn twice (build chariot (first random thing i clicked) in city 3), and save file example 45.2 (1350AD) after the bug (or what seems like a bug happened anyways), with city 2 i mean anyways now having 2 culture per turn on its monument (unlike city 3's monument that still has 1 culture per turn in save file example 45.2).