Check this out, and it is every time - zero plan for computer cities

  • We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

coodav

Banned
9 Badges
Jul 3, 2019
615
227
  • Age of Wonders III
  • Age of Wonders
  • Age of Wonders II
  • Age of Wonders: Planetfall
  • Age of Wonders: Planetfall Deluxe edition
  • Age of Wonders: Planetfall Premium edition
  • Age of Wonders: Planetfall Season pass
  • Age of Wonders: Planetfall Sign Up
  • Age of Wonders: Planetfall - Revelations
Just finished another game, and the computer's cities are just plain abysmal. This city is a good example, as it is crystal clear that the city has no plan:

Computer still dumb.png


Yes, that is a research specialization, with a level 1 hall, where there are NO LAND BONUSES. This is absolutely bonkers. Here is the same city a while later:

Computer still dumb 2.png


This is a city with a plan. Two energy sectors, ultimately being supported by a water sector, for maximum energy gain. Community wellness center for maximum growth and happiness bonuses. This will grow quickly, and pump that energy out.

Here are the obvious necessary improvements:

Computer still dumb 3.png


Last one would obviously be this:

Computer still dumb 4.png


After adjustments, this city is actually worth something. It will produce a TON of energy, and very quickly, probably a few hundred per turn, no problem.

The computer appears to have no idea how to do any of this. There is no possibility of this working out in their favor in any way, and the bonuses I inherited were worthless. ON TOP OF THAT, it took 11 MANUAL unstackable changes to get this city back on track.

1. First production set - this is to get initial defenses up, no initial changes to the sector.
2. Remove research specialization (cannot be stacked with any further improvements)
3. Remove research exploitation (cannot be stacked with any further improvements)
4. Place energy exploitation (cannot be stacked with any further improvements)
5. Place energy specialization
6. Remove production specialization (cannot be stacked with any further improvements)
7. Remove production exploitation (cannot be stacked with any further improvements)
8. Place energy exploitation (cannot be stacked with any further improvements)
9. Place energy specialization

Then at populations 12 and soon to be 16, it requires 2 unique visits that cannot be stacked to do the same for them. And it could be worse.

How is it that the game design allows for such abysmal planning? In this case, I ended the game with 30 cities under control, 10 of which I built. That means that I went through this exactly same thing 20 times before winning.

Worse, every possible system for accelerating this is stymied. So if you try to stack 2 deconstructions per turn, which cannot be hurried, it will do ONE of the two, and you will lose all remaining production. If you do just one, it will do the deconstruction, and still appears to give production to the next building in que. That is a stupid game design mechanic. So unless you want to lose 4 turns of production, you will not stack sector deconstruction. There is literally no effective way through this, you are about to do maintenance to every city you take over every couple of turns for probably the rest of the game.

Guys, this is abysmal. And it is EVERY SINGLE TIME. The computer's build orders make precisely no sense, and the system here fights you every single step of the way.

In the end, it is clear that the construction order run by the computer makes no sense. It appears that the computer simply cheats, and gives itself optimal bonuses, despite land values, plus a hefty multiplier. But why does everything in the build que fight you? It is clear that a lot of this needs to be revisited. These are basic game design decisions that can definitely be resolved. They need resolution.

And it isn't like it doesn't currently exist. Note this:

Build order 2.png


This will move up if you click on it.

Build order 1.png


But this one will not move up. Level 3 cannot build before 2. It knows that the build order is sequential. The game ALREADY HAS THE MECHANIC. It just somehow forgot how to use it.

This is bad design. No reason for this to be this way except that the game is not finished. It is clear that we need the developers to go ahead, take a minute, and figure this out.
 

Attachments

  • Build order 1.png
    Build order 1.png
    325,6 KB · Views: 0
Last edited:
  • 2
Reactions:
I agree AI needs an improvement in this but I suggest you to use a workaround - de-annex several sectors at once and then annex them back, it can save several turns.
I dislike "removing something takes exactly 1 turn" mechanic too.
 
  • 1
Reactions:
I agree AI needs an improvement in this but I suggest you to use a workaround - de-annex several sectors at once and then annex them back, it can save several turns.
I dislike "removing something takes exactly 1 turn" mechanic too.
Remember, you are still taking 2 turns to do it: turn 1: remove, turn 2: re-annex. This is 2 turn, also manual, to do the same thing. It can be stacked I suppose, so that is a bit of time you save if they are not in-line, where it would take more. I get that you can do this on top of what you are doing in-town, but it isn't a really awesome solution.
 
Last edited:
  • 1
  • 1Like
Reactions:
Do you still have the save from before you fixed up this specific Colony?
Sure do. Note attached.

Note that this is the earliest I could go. I have it only save 20 saves, which I think is default. That is the oldest auto-save I have, after I did the psi-fish thing. I would love to give this to you earlier, but this is the best I can do.

Edit:

And Jordi, thanks for looking into this. I really like how this game works, and would love it if it were smoother. This is a big deal.

Edit 2:

Note that I did this for a lot of cities. Because I had like 10 or so turns with the cities by this point, some of them were already well on their way to being renovated. That said... oof. This was a slog.
 

Attachments

  • AUTO SAVE - Altanan Hartheka, Turn 85.ASG
    1,1 MB · Views: 0
  • 1
Reactions:
Let me also give you the winning save. Note that the computer will be taken out by the Therians if you hit 'end turn.' I have their capital, and they apparently dispatch the leader after my move. That said, you can see how much went into revising these cities before the end.

Now I always tend to mind the cities when I capture them. I typically play with more AIs, and you can't let your guard down ever with their bonkers production, so I take the effort to win right. Either way, you should be able to see that the cities I got, and the cities I ended with share little in common.

Ultimately, I do wish I could go back further than this to show you exactly how bad it was. Maybe you could see this in action by running a computer against itself?
 

Attachments

  • AUTO SAVE - Altanan Hartheka, Turn 103.ASG
    1,2 MB · Views: 0
  • 1
Reactions:
Oh! just look at the cities from the computer at the other end. Tons of stuff like this:

Computer still dumb 5.png


YAY! This one has a sector I don't have to re-build. That said, there is a research sector on a 1 research plot, and another on a 0. Just... yeah... room for improvement for sure.
 
  • 1
Reactions:
Would you been so cool as the AI you wouldn't need a plan too. But basic economy and logistics knowledge would benefit the AI a little :p
Yeah, I can only wish I was as cool as the computer was. My only regret is that I didn't get to know him better.
 
  • 1
Reactions:
I find this hard to judge without seeing the surrounding land. If you cannot find a spot to build a level 3 research sector, do you simply build none at all? As long as we are not seeing something like a level 1 food exploitation in a possible level 3 research sector and vice versa simultaneously, I think that's fine.
 
Check the saves out for yourself. They are there. The visibility for the Blue team is limited because of the fog of war. That said, the turn 84 post was 1/2 fixed and still terrible.

No matter what though, building a specialization on a 0 improvement tile is a bad call.
 
I find this hard to judge without seeing the surrounding land. If you cannot find a spot to build a level 3 research sector, do you simply build none at all? As long as we are not seeing something like a level 1 food exploitation in a possible level 3 research sector and vice versa simultaneously, I think that's fine.
Energy exploitation is the only exploitation worth building in a case where you aren't able to find a matching sector for days.
 
As this has drawn some attention, I'd like to add something. The AI doesn't really explore the exploration sites in my games, and I've run into this consistently. Maybe they weren't coded to since they were visit sites before and they provided their bonuses without clearing them.
 
  • 1
Reactions: