I saw pretty much the same shocking under-utilisation of tiles.  The only difference between Intelligent and Ultimate is that leg-up in terms of abilities.

I'm going to cross my fingers and hope that limiting farms to one per planet will help the AI avoid stupid mistakes like these.

Ouch.  I just saw the spectacular farm on the Precursor Library trick recently.

I don't have any fancy pictures, and I don't know if it makes the "Best" category, but AI transport use deserves special mention.  Which planet to invade?  The one right next door? No way!  Let's go get the one across the galaxy!

Or, my favorite:

1 - Send fleets to destroy defenders.

2 - Send transport toward undefended planet, poised to invade.

3 - A defender pops up?  Oh well, send the transport back toward that other planet across the galaxy.

4 - Have fleets destroy the new defender.

5 - Where'd the transport go?

A sad result of no invasion priority and having the AI move transports before warships.

Lone transports are a pet peeve of mine.  When I go and invade, I usually (except that time when I was getting stonewalled by Yor defending troops) take enough transports to finish the job.  The AI sends its transports unescorted at a target without even considering how likely it is to get shot to pieces before it even gets there.  It only has some success with this strategy when there's a swarm wiping out defenders.

LOL, yeah i find great amusement in that all the time! Iv'e even checked out AI transport targets to see if they were targeting certain planets for any reason. High influence, econ or trade good, but ive never pinned anything down, seems random to me.



I actually don't think we need farms at all in the game. I think pop cap should be tied only to the planet class. In a GC3 i imagine, food for those planets would be tied into planetary trade.
Trade i think is a huge failure in the game, tying it into feeding your population would make it important.

I would have two types of trade.

One, between your own planets which is feeding your population, and,
Two, Trade how it is now, but the route limit would be based on influence, not an arbitrary number.

Anyway...i digress.....

Also in the poor war-fighting category: the tendency of some civilizations to send fleets from halfway across the galaxy to destroy an undeveloped starbase.

 

I can only imagine that the AI is virtually clicking on the starbase and thinking 'Zero defences?  Awesome!' when it does that.  Same with all the little scout ships that I parked in the shape of a smiley face.

The obsession with starbases needs to be worked out.  Whether it's sending fleets across the galaxy to pick off a worthless base or sacrificing half the armada against a well fortified base, the decision process needs improvement.

I have some others.

1 - The decision process for choosing targets needs help in general. Right now it's basically "is my fleet stronger?"  If yes, attack.  If no, find another target.  Sounds good, perhaps, until an AI sends its entire armada across the galaxy while a couple powerful enemy fleets pillage its core systems.  I mean, they fly their fleets right past the incoming enemy because they are individually too weak to stop them.  Two or three fleets in succession, though, could stop the threat.  Likewise, the powerful fleets ignore the outbound enemy because they are targeting the lone defenders.  Really? Our fleets just sail right by each other because we figure we'll have better luck against each others lightly defended worlds?  Oh boy.

2 - Speaking of lightly defended worlds . . . nuff said.  In my Kushana Initiative AAR, however, my conquest of the Yor systems was slowed significantly by large fleets of defenders.  Like 3 to 6 per planet.  Made for slow going after I captured their Orbital Command Center.  Highly unusual in my experience.  They didn't have much to do with their ships, so I guess they just let them pile up.  They could have done a lot of damage if target priority were better, as mentioned in 1.  Actually, they probably could have turned me back if they made a focused effort to stop my fleets and transports instead of . . . wait for it . . . sending their remaining fleets on a 10 week journey to my not-so-lightly defended worlds.  Sigh.

3 - The thing I was planning to post about in the first place.  I've seen numerous cases where an AI that has been slow to militarize is attacked by the local bully.  In my mod, the weaker AI is usually very quick to get rolling with weapons, but also usually makes the stupidest choice possible.  So the bad guy comes in with, say, guns and armor.  Does the newbie take advantage of that and go for missiles or beams?  Of course not.  Guns is the flavor of the game, so they start building fighters with 1 gun attack to counter the invaders coming in with . . . 2 armor?  That's not fighting uphill, it's grabbing a sled and hitting the slopes.  The much vaunted ability of the AI to adapt to what the enemy throws at it is sorely lacking.

Ok, enough for now.  I don't mean to simply beat up on the game, but these are some pretty spectacular failures that would need to be addressed for a GC3 to be an excellent game.

Well actually, if guns are the cheapest weapon to research and you can cram them on and swarm enemies with armour, it makes some sort of sense.  But otherwise, yeah it makes a lot of sense to pick something for which there isn't a counter yet.

You know I was thinking about this and maybe it would help to work up the pseudo-code to address these AI issues.

For example, taking the example of fleets or ships trying to determine which enemy fleet or ship to target, let's say that the AI has a list of enemy ships and its own list of ships.  It considers two fleets the same as one big group if they're within a week's travelling distance of each other.  Then all the fleets or groups are ranked by strength.

 

Find my first group.

Find the first enemy group.

While I have groups without new orders:

If enemy group is an unarmed scout, find next enemy group.

Else if my group is unarmed and unescorted, evade nearby enemy groups.

Else if my group can attack its current target this turn with a reasonable chance of success, attack now.

Else if my group is escorting transports to target or home, find my next group.

Else if the attack counter has been incremented half as many times as I have groups:

If the escort counter has been incremented a quarter as many times as I have groups:

If my group is a small fleet or ship, orbit one of my planets.

else send my group to take up position near my planets to wait for enemy groups.

End if.

If I have transports, send my group to escort transports and increment the escort counter.

End if.

Else if my Group has a comparable strength to Enemy Group, move to intercept and attack, and increment the attack counter.

Else if I have other groups available, issue a move order to concentrate my groups.

Else find the next enemy group.

Wend.

 

There's probably some obvious bugs with this, but you can see what I'm shooting at.

The willingness of the AI to trade away half their recently colonized galaxy for some techs and warships during a period where no warships are necessary (for instance, when there is no war). The AI needs a bit more intellect when it comes to valueing future gains.

The Troop Transport thing makes fighting an AI that is vastly, vastly more powerful than you not a lost cause, the AI is quite bad at exploiting crushing military advantages for actual territorial gain. As for AI troop transport preferences: I find that they have the tendency to try to retake planets they lost first, before invading other planets. Even when they lost the planet they try to retake a long, long time ago.

 

Edit: I'd like to stress, by the way, that inspite of its flaws, the AI in Galactic Civilizations II does a much better job than any other AI I can recall in any strategy game (or any game short of Chess) period!

Recently, in a modding effort of mine, I've seen multiple AIs decommission the starport on more than one planet near game start-in a gigantic all-abundant, no less.

And Neilo, it sounds like you'd prefer a model closer to SOASE's.  Have you played it?

Of course! I always have a game going using the distant stars mod!! I wouldn't say prefer though, GC2 is my favorite game including all it's shortcomings and flaws.

This, I've never seen. Do you have any idea what caused it?

Pretty sure I've never seen it, either, but I have seen the AI build over the Recruiting Center and occasionally economic structures.

I build over the Recruiting Center at times myself, if the planet's reached its full pop and I won't be launching transports from it.

All I can tell you is that I didn't change a thing in the improvements file.  It doesn't make any sense.