AI not expanding

[fab_floyd]

Hi guys

I´m trying to play this awesome mod again, but it seems to be that the AI doesn´t like to expand his territories even tho i´m playing at monarch dificult.

To test this problem i let the AI auto play 100 turns in blitz - monarch, but it barely seems to reach two cities

Is there any configuration game to sole this problem??

Also, always when i do an atack move, it doesn´t shows any animation (sames as if quick moves were enabled, but it´s not)

 

[Thunderbrd]

Hi guys

I´m trying to play this awesome mod again, but it seems to be that the AI doesn´t like to expand his territories even tho i´m playing at monarch dificult.

To test this problem i let the AI auto play 100 turns in blitz - monarch, but it barely seems to reach two cities

Is there any configuration game to sole this problem??

Also, always when i do an atack move, it doesn´t shows any animation (sames as if quick moves were enabled, but it´s not)

What version are you playing? Are you on the latest SVN?

 

[fab_floyd]

I´m playing 38.5 version

 

[Thunderbrd]

Yeah, there's a lot that's been done to improve on that through the latest SVN. We have a few more things to do but v39 is approaching pretty quickly I think.

 

[fab_floyd]

Downloaded the lastest SVN versión (10487) , made an autoplay game: monarch - ultra fast (100 turns)

AI reaches early ancient era, max 4 cities founded (average of 2) , a lot of enforcers

Is that normal??

 

[raxo2222]

Downloaded the lastest SVN versión (10487) , made an autoplay game: monarch - ultra fast (100 turns)

AI reaches early ancient era, max 4 cities founded (average of 2) , a lot of enforcers

Is that normal??

I guess Monarch always was easy, try Emperor or Immortal.

Ultrafast is test speed - it may distort AIs.
Do autorun on Blitz speed.

100/500th is 20% of game, tech leader should be on verge of reaching Medieval era.

 

[Thunderbrd]

Ultrafast is test speed - it may distort AIs.

The way it would distort things is that you can only have one tribe at a time so the time it takes to get the city into position is relatively a lot compared to slower gamespeeds since the movement rates don't scale. I think I might be hard pressed to have more than 2-4 cities by early ancient when the game is progressing that quickly.

There are also a lot of buildings that come up around the time you can start to grow and expand so that could be something the AI is delayed by juggling some. Not sure without looking much closer.

Right now I'm about to place a 3rd city in a test game on immortal/epic and I'm not quite up to sed lifestyle yet but it's approaching quickly. We'll see if I'm up to 4 cities before that point or not. Whisperr and I are playing a team game here and unfortunately, since the last AI settings adjustment, we're already ahead of them in tech so they are running behind us there (at least the ones we know are) so its hard to tell how well they are growing in comparison but the next two most advanced AI have 2 cities at this point. They seem to be expanding at the rate they should since they pretty recently got tribalism, but its unfortunate how much we had to reduce their construction cost edge - it's hurt their progression a lot.

Still, they'll start getting a competitive edge for being behind after Sed Lifestyle since we have WFL and tech diffusion on and no handicaps for humans. So if they do fall behind, they'll be all the more lethal in a way. Will be a good test to see how much it really helps them to stay competitive.

One really neat thing that happened was that the most advanced neighboring AI did put together a small war stack to attack me earlier. Quite hilariously, he ended up attacking my wife's ambusher on a forested hill as he tried to approach my borders and was accidentally repelled so the war never began and he wisely gave up on the effort thereafter since he soon got tribalism and shifted to growth mode.

 

[raxo2222]

They seem to be expanding at the rate they should since they pretty recently got tribalism, but its unfortunate how much we had to reduce their construction cost edge - it's hurt their progression a lot.

AI construction costs are same, its tech cost ladder, that was flattened, so now its 3 per step instead of 5 per step.

 

[Thunderbrd]

AI construction costs are same, its tech cost ladder, that was flattened, so now its 3 per step instead 5 per step.

OK, I thought it was the construction rates that were adjusted. Anyhow, point being, I can really tell the difference. We might need to go to a 4 per step on that but I'm wondering if others are finding the AI too easy as well.

I'm also finding that gold has been a total non-issue. It's back to a total glut of gold again. Looking at exactly why, I'm thinking we need to greatly increase base civic upkeep costs on the first rung(s) of civic advances. They don't have to be so inexpensive really. Buildings are mostly reasonable and it's understandable that Tribal Warfare would give enough 'free units' that I'm not getting hit there. So something else needs to cause a little gold pain here. Tribal warfare also doesn't seem to be needing the +gold modifier it's getting but that's actually a relatively minor adjustment at this point so it's more just the overall. I can only speak for the point I'm at but +40 and over gold per turn before tribalism without any help from traits to achieve that (and not many penalties either) was a bit much. I'm not quite sure why but it appears that even 'low' upkeep civics are coming out at 0 - might be because I do have a -10% modifier and that low upkeep is only amounting to 1 - 10% bleeding down to 0. Certainly it should be costing a bit more considering the civic steps up are significant improvements now. I like a lot of the adjustments that have been made there to make that clear.

Spose this is not the thread for that conversation.

 

[JosEPh_II]

I can only speak for the point I'm at but +40 and over gold per turn before tribalism without any help from traits to achieve that (and not many penalties either) was a bit much.

Why?

As you have given no context of the game in terms of GS, Difficulty, # of AI, or Map size. All very pertinent details as to your immediate experience.

 

[Thunderbrd]

As you have given no context of the game in terms of GS, Difficulty, # of AI, or Map size. All very pertinent details as to your immediate experience.

I did mention the two primary details, actually, not all of those but to review:
GS: Epic
Difficulty: Immortal
# of AI: 12
Map Size: Large

The overall answer of why it was a bit much is because I'm on 100% research and still have quite a few buildings I can construct to produce more gold. There is no gold strain whatsoever at this stage. Hitting tribalism I have a few thousand gold. Everything is upgraded. I did see it go down to +twenty something after my first city was initially planted but it's already recovering a lot. My wife saw her gold go up after the first extra city was placed. Her gold is more than mine and she can build even more buildings to get more still.

 

[JosEPh_II]

At 40 gold/turn at Tribalism with 3 cities (and you had to capture 2 of them because you can't build cities Until tribalism) is not excessive. Owning 3 cities before tribalism is. And also a reason you are shooting ahead of the AI in Research. And other areas too.

How many AI captured cities before Tribalism? I bet none.

The real game twister is taking cities too early in the game, ie before Tribalism. That skews Everything because you multiply all your assets with every city taken.

If Civics would make you lose money because you have 3 Cities before or right after Tribalism the AI would stagnate again. We walk a tight line here.

But then Civics are not a finished product yet either.

 

[Thunderbrd]

At 40 gold/turn at Tribalism with 3 cities (and you had to capture 2 of them because you can't build cities Until tribalism) is not excessive. Owning 3 cities before tribalism is. And also a reason you are shooting ahead of the AI in Research. And other areas too.

I didn't have any cities at that point. I said I'm about to have a 3rd city now. Been running at +40gold/turn where ideally, I think it should be around 0 at 100% research. From there, each city should cut in a little bit. I like to be forced to downshift the slider to afford upgrades.

I suppose it doesn't have to be THAT painful but having the civics cost something in civic upkeep should help. I do have a -10% modifier so maybe without it I'd be at 1 gold for the upkeep on those civics that are at 'low' upkeep. Do you know where the upkeep cost base is set and how that's done exactly? I've never really looked into it too deeply and I think that just a little bit more cost on those would help with system balance. I'm not sure what the end formula for civic upkeep actually is derived from and what all factors play into that but it seems that even game-wide, it's a little weak for the kinds of inflated amounts of gold we get in C2C overall.

Whisperr actually gained gold for some reason when she placed a city but I think it was due to her city maintenance amounts being so negative that new cities represent gains... I might need to make that minimum out at 0 upkeep rather than being possible to profit from more cities.

How many AI captured cities before Tribalism? I bet none.

None I can see but one gave a go of it. It's not Deity level though so they didn't have extra settlers to work with.

If Civics would make you lose money because you have 3 Cities before or right after Tribalism the AI would stagnate again. We walk a tight line here.

The AI should be able to spread at less than 100% research. There's a global that says how much they should be allowed to tolerate being set to gold on the slider without being in the negative while still allowing further city founding. I think someone might have set it to 90% which is probably too high I think.

 

[raxo2222]

The AI should be able to spread at less than 100% research. There's a global that says how much they should be allowed to tolerate being set to gold on the slider without being in the negative while still allowing further city founding. I think someone might have set it to 90% which is probably too high I think.

Isn't this entry?

    <!-- AI Gold Slider Safety Percent begin -->
    <!-- This is the BASE amount of slider allocation required towards gold that the AI feels comfortable with -->
    <!-- It noteworthily affects how much tolerance the AI has where it feels ready to grow -->
    <!-- Reduce to make the AI more sensitive to gold concerns and increase for more reckless behavior -->
    <!-- This may be rather interesting to convert to an AI personality matter or at least give a personality based modifier eventually -->
    <!-- Original Value was set to 50 - I've adjusted to 40.-->
    <Define>
        <DefineName>SAFE_GOLD_PERCENT</DefineName>
        <iDefineIntVal>40</iDefineIntVal>
    </Define>

So higher value could make AI more expansive.

 

[Thunderbrd]

Isn't this entry?

    <!-- AI Gold Slider Safety Percent begin -->
    <!-- This is the BASE amount of slider allocation required towards gold that the AI feels comfortable with -->
    <!-- It noteworthily affects how much tolerance the AI has where it feels ready to grow -->
    <!-- Reduce to make the AI more sensitive to gold concerns and increase for more reckless behavior -->
    <!-- This may be rather interesting to convert to an AI personality matter or at least give a personality based modifier eventually -->
    <!-- Original Value was set to 50 - I've adjusted to 40.-->
    <Define>
        <DefineName>SAFE_GOLD_PERCENT</DefineName>
        <iDefineIntVal>40</iDefineIntVal>
    </Define>

So higher value could make AI more expansive.

That's it. So it's been set back down. You don't want them to be overly reckless in either direction. They can choke off their research entirely by overgrowing and they can fail to invest enough into future research by underexpanding. I think the 90 I remember being set there was in the GEM scenario, not core.

 

[raxo2222]

That's it. So it's been set back down. You don't want them to be overly reckless in either direction. They can choke off their research entirely by overgrowing and they can fail to invest enough into future research by underexpanding. I think the 90 I remember being set there was in the GEM scenario, not core.

You can't set anything like that in scenario, though mapmakers may provide modified files with maps, just like I have latitude/longitude rescaling/removing modmods for spacemaps.

 

[Thunderbrd]

You can't set anything like that in scenario, though mapmakers may provide modified files with maps, just like I have latitude/longitude rescaling/removing modmods for spacemaps.

He does.

 

[raxo2222]

He does.

Can I see that scenario? No idea how scenario file itself can override any globals.

 

[tmv]

The AI should be able to spread at less than 100% research.

How much does the slider really mean? Ever since the => change the sliders haven't been the same.

 

[Thunderbrd]

Can I see that scenario? No idea how scenario file itself can override any globals.

I am saying last I knew he was including files for use with it. Its Pit's massive GEM scenario. One of the most active threads in the maps subforum here.