Are the AI upgrading luxuries/strategic resources after March patch?

[Socrates99]

I have the mode on in my current game and John Curtain has 2 olives and 3 dyes in the trade screen. So I panned over his empire and there's two dyes improved between Maitland and Toowoomba. Not saying there isn't a problem though. He's got cocoa and spices he hasn't touched and at least 4 more olives that'd help him get a monopoly.

Looked some more, Ambiorix has 4 whales and 2 ivory improved. Poundmaker has 2 jade. Again, lots they haven't improved, especially Ambiorix who has a pretty good sized empire.

 

[AntSou]

AI not improving enough resources is nothing new in my experience. The reports after the Monopolies update was that the AI was barely improving them at all. Most of the screenshots/reports in this thread look "normal" to me.

I've always seen it as a consequence of the AI's poor handling of builders. Not just ignoring luxuries, but ignoring pillaged tiles and building too many farms, etc.

 

[Tigranes]

Yes, without Monopolies.

There you have it.

 

[kaspergm]

AI not improving enough resources is nothing new in my experience. The reports after the Monopolies update was that the AI was barely improving them at all. Most of the screenshots/reports in this thread look "normal" to me.

I've always seen it as a consequence of the AI's poor handling of builders. Not just ignoring luxuries, but ignoring pillaged tiles and building too many farms, etc.

I disagree strongly. There’s been widespread reports from many different and independent players of a significant shift in AI behaviour since release of M&C. Writing it off as “just normal behaviour” seems a tiny bit arrogant to me, sorry to say.

 

[AntSou]

There’s been widespread reports from many different and independent players of a significant shift in AI behaviour since release of M&C.

"The reports after the Monopolies update was that the AI was barely improving them at all."

So we're not disagreeing on that.

Writing it off as “just normal behaviour” seems a tiny bit arrogant to me, sorry to say.

I'm not writing it off, I'm saying that I see no difference between the described behaviour in this thread and my experience pre-Monopolies update. I'm not claiming that's the case, only that I can't perceive a difference. Why is it arrogant to share my experience? Since there's a big difference between the reports in this thread and the one reported after Monopolies, I'm suggesting it might be back to "normal", as in, as poorly as before.

This was posted by @Pfeffersack July last year:

Spoiler :

https://forums.civfanatics.com/thre...t-improving-ressources-over-long-time.660377/

If someone were to post this today, it would be used as clear evidence that "AI still not improving resources with Monopolies mode." So better not to jump to conclusions.

This has literally been one of my main gripes with the AI since ever. They'll often not improve duplicates, even city-states, and then not fix the tiles if they get pillaged. This thread so far has not clarified whether the problem reported after Monopolies persists, or if it's back to "normal".

Finally, this being a Mode which gives an extra improvement to place on an improvement, if the AI already struggles with the first step, it is to be expected it will perform even poorer when given an extra layer of complexity. So even without any bugs, AI will likely still do worse with Monopolies Mode than without it.

 

[Abaxial]

I disagree strongly. There’s been widespread reports from many different and independent players of a significant shift in AI behaviour since release of M&C.

You just have to compare the experience of trading before M&C and after it. It used to be that the most common trade was one luxury for a different one. Now this is relatively uncommon, because AI civs usually don't have duplicates of any luxury. The question is whether the change in behaviour after M&C was a bug or a deliberate design.

 

[Guynemer]

I am playing a game without M&C and all of the AIs have plenty of extra luxuries (5+ is common) to trade.

 

[Tigranes]

You just have to compare the experience of trading before M&C and after it. It used to be that the most common trade was one luxury for a different one. Now this is relatively uncommon, because AI civs usually don't have duplicates of any luxury. The question is whether the change in behaviour after M&C was a bug or a deliberate design.

It was explained very clearly what is going on. AI assigns higher priority to creating Industry/founding Corporation and does not care so much for improving luxuries until it is given a tech that allows to do it. Economics tech is the key to understand the root cause of the bug. The moment you artificially move Corporation to Currency -- things go back to normal, that's the only thing that mod fixing the issue does. No need for more speculations and investigations, everything is established but not fixed by devs.

By the way even with a fixing mod I never ever seen an Ai creating a Corporation. And Corporation mode is the only always-on mode in my games.

 

[Stringer1313]

Shouldn't AI be improving all possible luxuries even with Corporation mode turned off? What is the reason behind not wanting AI to improve too many luxuries? For trade, etc.?

 

[Pfeffersack]

It was explained very clearly what is going on. AI assigns higher priority to creating Industry/founding Corporation and does not care so much for improving luxuries until it is given a tech that allows to do it. Economics tech is the key to understand the root cause of the bug. The moment you artificially move Corporation to Currency -- things go back to normal, that's the only thing that mod fixing the issue does. No need for more speculations and investigations, everything is established but not fixed by devs.

By the way even with a fixing mod I never ever seen an Ai creating a Corporation. And Corporation mode is the only always-on mode in my games.

I decided to test this theory without using a mod - by starting a game with a modern era advanced start (= everyone having access to Corporation)...and I didn't need to play really long to see a clear difference...

Spoiler :

I think that speaks for itself...but even if it is only lacking priorization which keeps the AI improving (more) luxuries until Corporation - a fix is needed, as missing out the benefits of both trading away excess copies and of early industries is just a huge, unnecessary AI flaw.

EDIT: Save added.

 

[Melkus]

And here's a resource report from a recent game with the little mod I mentioned earlier in the thread. The AI does fine in M&C Mode when Corporations is moved forward to Currency. ( No, no one has researched Economics yet.) They're not perfect, but being ineffective with resources is on the same level as many other AI flaws, while the problem in M&C without the mod is crippling due to a lack of trade and amenities. (No idea what is up with Harald's 2/4 "monopoly" in Jade)

 

[Pfeffersack]

In a normal none-advanced start game I also don't see any real improvement after they have researched Corporations. Not sure what is exactly their problem...I would probably (still) built less Spaceports and GDR, if I were in the AIs place, but I see AI workers running around...they just don't get used to improve ressources (Luxuries worst, but even strategic ones are neglected) as it seems:

Lacking priority when it comes to building workers might be a part of the problem, but I have now cought the AI with an example where it even ignores a luxury, where one of its worker is already standing on top...

https://forums.civfanatics.com/thre...proving-resources.666943/page-2#post-16067231

I think more proof of an issue/bug is just not possible and and so I deciced to put that example (with pictures and a save allowing to reproduce the behaviour) in the corresponing bug thread.

 

[AntSou]

There are four city-states with luxuries within their borders in my game and none have improved them. Renaissance Era.

 

[Abaxial]

I don't think it is a bug, I think it is deliberate to stop the AI winning every game.

 

[Pfeffersack]

I don't think it is a bug, I think it is deliberate to stop the AI winning every game.

Beside that I don't think this is the case (why does it improve luxuries like usual in an advanced start game with C&M mode activated, as shown above?), can you elaborate why that would make the AI impossible to defeat? And on which difficulty? Deity, Prince, Settler? Honestly, given all the boni the AI in Civ6 needs to be competitive I don't think there is a any need to dumb them down on purpose (and I also doubt that Firaxis does spend coding ressources on this, because such a different extra behaviour would likely have to be "teached" the AI as well)

 

[Abaxial]

Because the "accidental" CV obtained early on from monopolies would be obtainable by the AI. So there would be a good chance of losing to an AI you never even met that managed to get good monopolies.

 

[Pfeffersack]

Because the "accidental" CV obtained early on from monopolies would be obtainable by the AI. So there would be a good chance of losing to an AI you never even met that managed to get good monopolies.

I get that the monopoly effect on tourism is still a problem in the hands of the human player, but how likely is such a scenario with the AI? Is one monopoly enough to trigger such a "en passant" CV (asking because AIs getting several ones seems even more unlikely)? Also, if that is true, I wonder if that problem occurs if the mod recommend here is used: https://forums.civfanatics.com/thre...after-march-patch.668702/page-2#post-16061065
Anyone having played with it and got a surprise culture victory of the AI? Also, this would be probably again a matter of difficulty levels. At best, I see that on Deity or maybe Immortal such an effect could turn the normal AI advantage with culture/tourism into a victory for them; below that the AI is likely overall too weak to pull it off that early. And then a deliberate nerf would mean that Firaxis has done this and invested extra energy to keep the high levels winable (which few players play anyway). Doesn't sound very likely for me, but ok - we can't rule out your scenario and I'm curious to see what the future brings - hopefully a fix to this AI flaw and a better balance monopoly/tourism formula

 

[Melkus]

Anyone having played with it and got a surprise culture victory of the AI?

I have played several games with it (mostly immortal) - in an average sized game (6 or more players), I don't think the Ai can ever win by monopoly (at least not so early that it feels like a premature end). The AI doesn't actively go for monopolies anyway, and they're still inefficient with their builders. And, the culture win math gets harder for everyone because more resources are developed.
The most important difference is that the AIs develop normally because they generate more amenities, trade income, industry yields - maybe even some monopoly money.

I still think the tourism multiplier should only be applied to civs that lack the resource. That part seems obvious to me. If I have wine already, why would my people travel to a country that has a lot of it? And the civ multiplier in the formula should go away too.

 

[Abaxial]

If I have wine already, why would my people travel to a country that has a lot of it?

Well ... England has wine, but it is expensive and not as good as French or Italian!

 

[Socrates99]

I have played several games with it (mostly immortal) - in an average sized game (6 or more players), I don't think the Ai can ever win by monopoly (at least not so early that it feels like a premature end). The AI doesn't actively go for monopolies anyway, and they're still inefficient with their builders. And, the culture win math gets harder for everyone because more resources are developed.
The most important difference is that the AIs develop normally because they generate more amenities, trade income, industry yields - maybe even some monopoly money.

I still think the tourism multiplier should only be applied to civs that lack the resource. That part seems obvious to me. If I have wine already, why would my people travel to a country that has a lot of it? And the civ multiplier in the formula should go away too.

Same reason Californians appreciate French and Italian wine or American tourists go to Belgian or German breweries. Once a nation/culture gets known as THE place to get a certain commodity people view other sources as imitations or knockoffs even if the quality is on par.