A Better AI.

[MikeH]

Yeah, I find I have to beeline to liberalism to guarantee getting it now.

[Common Sensei]

Originally posted by Arrian
I like to try the "Democracy slingshot" by researching printing press and constitution before liberalism and then taking Demo for free (typically nabbing economics too for the free GM). This has been more difficult since 2.08, and it would seem even moreso with Blake's changes (I've not been trying out the better AI version-by-version).

-Arrian

I prefer taking Astronomy as the free tech, as it is one of the most expensive techs you can get for free and it allows you to build another science building right away.

It's like an Olympic Triple Long Jump of research.

[darrelljs]

Edit: removed redundant question.

Do you think at some point it will be necessary to remove some of the hidden AI bonuses? For example the low cost to upgrade troops, the large maintenance discounts, and the bonuses against barbarians? These don't really scale for difficulty level and at some point might become obsolete.

Darrell

[foudres]

Originally posted by MikeH


I like to do that, if I can.

On a touch game i have to go liberalism first and will go for printing.

I can't afford to research no needed tech like nationnalism or banks so will not go for democracy or economics.

And printing is so good that it increase you research a lot so that not a bad deal.

On prince level (or monarch prior to Blake AI) i can afford researching side techs. Here it's not possible.

On the AI topic blake :

I go nearly allways for industrialisation goal on late game so i have thoses 20 power riflemens and i can make factory anywhere. I tend to see that even if the AI is on part on research it tend to go for it very late.

So i have a long time when i am the only one with 20 power units and with factories. I rush factory as fast as i can while pumping units.

At soon as i have done that the AI is basically so easy to defeat... Even when they have the tech they do not make enough factory and i have at least twice the production of my best oponent. I can make all the unit i want and destroy everyone that is still here.

So maybe there is better thing than this strategy... But i think that the choices of the AI for tech in the late stage in particular are not really good.

Does the AI aim for a far away goal like researching industrialism ?


Because my strategy for reseach is that i have key thing to have as soon as possible... And this drive way i research

at the beginning :
- research bronze working
- research tech needed to improve the title i have. But i will not try to research calendar if i do not have a lot of calendar resources trough (I will trade to have it).
-research iron working if i have some jungle.

- research the tech that allow tech exchange

- go for CS

-go for Liberalism

- go for banks

- go for rifles

- go for industrialism...

It's just an exemple i do not always do that... but at least i want to do something (a goal) and for that i will reseach some specifics things.


Like paper give me nothing... And gunpowder is useless... But paper is needed for liberalism and gunpowder is need for riflesman.

Maybe i'am wrong but by reading the thread i understand the AI will search a tech if it want it and it's available for research...

But if a very good tech is behind a tech the AI do not really want to have.... this good tech will only be owner by the AI after trade or something.



Anyway that my 2 cents.

[DaveV]

Blake:

Here's a case from my latest game where I think the AI picked the wrong city site. The body of water to the north is salt water, so building where my axeman is standing would allow construction of a lighthouse. Even if there's a food special on the fogged square, I'd still build on the axeman square (and build another city to use the food special).

I was a couple turns from settling at the axeman site, and was hoping Caesar was moving to a distant location when he passed it by. Now he's spoiled the site I wanted, and I'll probably have to raze his city.

Edit: changed harbor to lighthouse. You know what I meant.

[Kuciwalker]

On the other hand, moving northeast would put three more water tiles (with no bonuses) in the fat cross.

[Blake]

That is strange placement from Caesar.

I'd appreciate the save, or if you don't mind, a world-builder screen shot since I can't see why he wouldn't choose that GL hill unless there is something he can see but you can't.

It is possible he was just tempted by all the forest (especially if Agustus) but I find that hard to believe it'd be enough to cover the +700 something from the 4 lake tiles and the coastalness.

I could also extend the anti-deadlocking code to include coastal tiles in some fashion so it tries to keep them free. That code is somewhat expensive to run but there are ways around that...

[DaveV]

Originally posted by Blake
I'd appreciate the save

On the turn of the screenshot:

[DaveV]

This one should be before the city was built:

[Blake]

Thanks. It's actually a bug in Vanilla, when Iustus performed the original merge he replaced the warlords code (which doesn't exist in Vanilla) with something similar which does exist in Vanilla. Unfortunately that being the function which mistakenly identifies lakeside as being coastside. I'm surprised it took this long to get caught, I guess just not many people use the Vanilla version.

[DaveV]

Well, here's one guy who's still playing vanilla, and really appreciates your efforts.

[Blake]

I also generally boosted the value of those special lakes, making them recognized as a solid food source.



Thanks.

[Zuul]

What is this CFC thread your are talking about?

[Iustus]

Originally posted by Zuul
What is this CFC thread your are talking about?

Ask and ye shall recieve: CFC Better AI thread

-Iustus

[Saurus]

Played a few games again...

I was just wondering a bit about holy cities.
Now that A.I is happily spreading every religion so that they spread like a wildfire, the holy shrines seem to be very, very powerful now considering you hardy even need to build any missionaires yourself. Maybe even a bit too much so!

Normaly, before the A.I patch, by the time I have my holy shrine built, I´ll get out some 4 - 12 gold/turn from it initially. Then gruadually increasing.

Now with the latest patch by the time I have built my holy city I have a MUCH greater income. During my first game I had an income of 26 initialy and during my second a whopping 32(!)
This modified with the economic buildings makes for a huge difference compared to how the game was playing out before the patch.
I think this is more of an issue on higher levels (Since I assume religions then spread quicker becouse of A.I building missionaires much quicker)

Don´t now quite what to do about it though...
The feeling I´m having right now, though, is that shrines are way too powerful as it stands right now.