1412 testing

[gdwitt]

Played game in latest patch. 1.412 Emperor Denmark Standard 10 civs 13 CS.
It's a dramatic change from 1.392
Little wonder competition except for Darius , who ended up getting eaten by others, and Hiawatha.
No civ had excess gold. All had around 200-500 at each check except Japan which wasn't doing much in its corner.
Rome did it's typical bad boy act.
Little "Small Civ" competition from Germany or Russia like usual.
I excluded Greece from game.
The highest bid was 60 from AI.

I was far ahead of AI in nearly all aspects.
Only 2 civs were keeping ahead of my science and I wasn't trying to hard here.
The others did not lag however.

Overall, it seems too easy. Much easier than 1392 or earlier Emperor versions.

No bugs except that I didn't receive any castles for the liberty finisher. I had already researched chivalry. This was also a bug in 1392. It's not a bother with ingame editor or if I'm winning by such a large margin.

Are the standard maps bigger?

How to get AI to use its Iron? I used Ingame editor to give each civ 4 Iron sources. By turn 150, no one had more than a few swordsmen. Most of civs using levies, archers, spears. Their workers are more focused on building 3 food farms than mines on iron.

 

[Thalassicus]

I'm glad to hear excess gold is no longer an issue. VEM uses vanilla map sizes.

Every player has precisely 5 iron in their territory at the start of the game; adding four more would bring the total to 9. Peaceful AIs do not prioritize strategic resource units like swordsmen. It's the militaristic AIs who generally build strategic units.

 

[The_J]

Moderator Action: Moved to the main forum.

 

[mitsho]

My experience tonight was pretty similar. There was quite some competition regarding science and military. Even expansion, now we just need to make them cross the seas

It is annoying however to have the lone AI settler slip through the gaps between your Tall cities.. Cultural expansion of the tiles seems extremely low. That effect is a bit late on the finisher. Maybe switch it with the happiness from the first policy on the right, one normally doesn't need happiness that early...

I've played until the middle of the middle ages, I do need to play an advanced start to see the later eras (since normally I don't have time to finish games and then there's already a new version out ).

One other thing is City States:
Should they have Swordsman/Strategic units? That makes them quite strong at conquering lonely cities. Do they have a cap on their strategic units
Have they become more aggressive? They went heavily after barbarian camps in my game so that no real barbarian pressure could be started (it was continents plus, small size, low sea, but there were many cs on the mainlands).
Maybe they should start with fewer units?

Btw. it was/is an emperor game with me as Arabia, Bismarck and Hiawatha on the continent. Hiawatha was slow at first in expanding, and then suddenly very fast, but then he was attacked by Bismarck who took one city and encroached on the capital pretty far. I was behind Hiawatha and he attacked me afterwards with a good mixed army (that has no chance due to the terrain and him sending the catapults first, but his horsemen are devastating to a degree ), despite I having allied with him and "attacked" the Iroquois.
On the other continent, Augustus and Napoleon were attacking Pacha and Elizabeth respectively and by the time I had explored the coastlines of the continent, England was conquered and Pacha reduced to one or two cities.

So, I'd say the next step would be to buff Tall civs imho and/or make some civs not attack in the Ancient era (to avoid repeatings and predictability).

 

[Thalassicus]

Sounds like an exciting game! If citystates don't upgrade warriors to swordsmen, what should we do with their warriors?

Some are saying expansion is too slow, others saying it's too fast, so I think I'll leave it like this for a while.

make some civs not attack in the Ancient era (to avoid repeating and predictability).

I don't believe we can do this part.

 

[mitsho]

Let me clarify:

Expansion of number cities was fast (or first slow, then fast), and this is good.
Expansion of tiles belonging to one city seems really really slow, but after one game, I'd be wary to make a judgment.

My "make them attack later" was in the same direction as "make them cross the seas", not really serious. That's more core AI behaviour... Wishful thinking so to say

And I'd say it's ok if they have one swordman, but in this game, Quebec took Cologne from Germany with two catapults and two swordsman. They've sent out swordsman taking out barb camps, it's just too much. Maybe just delaying the upgrade helps or not all units have to upgrade immediately... (I mean I can't do that either...). Or rather, if the problem is really them going too strongly after barbarians, they can keep their swordsmen

 

[Thalassicus]

What's your typical proportion between expanding with vs ? I think about 75% of my expansion comes from culture, and 25% from gold.

 

[Txurce]

I'm playing Korea, and am on a small continent with Egypt. In the low 100s (Medieval) Egypt suddenly trumped me four straight times for control of the lone CS on the island: 382/285. This may not be wildly off, but it's certainly not the AI backing off when topped on consecutive turns. More problematic, the next turn t signed an RA with me (300g), while raising its CS edge uncontested to 519/283. Needless to say, its gold isn't at that level: 237g/41gpt. (And the numbers went up again the next couple of turns. The algorithm needs some tweaking.)

 

[Thalassicus]

At least it's not in the 1,000 range anymore; getting very close now.

 

[Txurce]

At least it's not in the 1,000 range anymore; getting very close now.

Don't count out Ramesses!

 

[Seek]

I must say, the CS revamp is what I'm looking forward to most in G&K.

 

[Thalassicus]

Txurce, what were the circumstances of the spending battle with that citystate:

• how many tiles was the CS away from Egypt
• did Egypt know any other CSs
• militaristic or maritime/cultural
• hostile or not

Also, if you still have it, what did that time period look like in lua.log?

 

[Txurce]

Txurce, what were the circumstances of the spending battle with that citystate:

• how many tiles was the CS away from Egypt
• did Egypt know any other CSs
• militaristic or maritime/cultural
• hostile or not

Also, if you still have it, what did that time period look like in lua.log?

Rio de Janeiro (Maritime, neutral) is 7 tiles from the nearest Egyptian city, 11 from the capital. It's now 6 away from my nearest city (which was built later).

Rio was Egypt's ally. I took it from Egypt. Egypt did nothing for a while, then flared at essentially the identical point in which it happens in my games.

We are at the south pole, Egypt has one coastal city, and I haven't seen a trireme pass by. But the map is Continents Plus, so someone may have sailed down from the east. (My guess? No.)

What do you mean by "what did that time period look like in lua.log?" What it says in the Lua Console/Main State?

 

[Thalassicus]

Right, the "Ramesses X spent on Y influence with Z" stuff. I'm wondering if it was my code that did the purchases, or vanilla's code.

 

[Txurce]

Okay, in two turns where Ramesses kept spending, I see that the Mongols spent gold upgrading units (although I haven't met them) - don't see what anyone else did (haven't met anyone other than Egypt) - and can't see what Egypt did.

Am I missing something?

 

[Thalassicus]

The log shows every purchase the mod makes for an AI (in contact or not). If nothing shows up in the log, it's probably vanilla making the purchases. For the next version I've added "total AI gold" to the purchase lines so we can double-check they're not spending more than they have.

 

[Txurce]

Interesting. I also wonder why Mongolia was the only one doing anything (in this case, upgrading) for three consecutive turns.

 

[Thalassicus]

You mentioned their gold per turn was in the 40 range, so it would take at least 5 turns to buy a granary/stable for 200. If the vanilla AI spends any gold during that time it takes longer. In early eras I typically see one AI purchase every 3-5 turns.

Could it be Egypt traded resources with someone for gold, then vanilla immediately spent it on citystate influence? There will inevitably be odd behavior when dealing with AI decision-making. I'll be happy once we can make those incidents rare, instead of the norm.

 

[mitsho]

What's your typical proportion between expanding with vs ? I think about 75% of my expansion comes from culture, and 25% from gold.

In a typical city, at this point I buy 1-3 tiles. This includes tiles towards other civs with ressources and natural wonders or tiles with strategic values. Afterwards, there's no need to buy a tile unless a settler is trying to sneak by and I'd rather use the gold for buying buildings or units that I need.

We can build culture in a city. What if that is available earlier and also lowers the threshold of tile aquisition so that it'd be really strong to quickly expand without investing money?

 

[Thalassicus]

I'd completely forgotten about the culture process... shows how often I use the processes. Maybe make it available at Philosophy, and the gold process at Trade?