• 📚 A new project from the admin: Check out PictureBooks.io, an AI storyteller that lets you build custom picture books for kids in seconds. Let me know what you think here!

"AI" stands for . . .

Bushface

Deity
Joined
Nov 28, 2005
Messages
2,550
Location
Torquay, England
. . . Absolute Idiocy, at least sometimes in respect of worker activities. Currently I have three vassals; between them, I can see the following examples of less-than-optimal behaviour :-
1). 4 workers in a fat cross; one building a Farm, one a Workshop, the other two building another Workshop on a tile which already has a Farm. There are two undeveloped grass tiles in the same fat cross.
2). A worker building a windmill on a grass hill. All other tiles in that fat cross are undeveloped: there is copper on one grass tile, and several riverside grass tiles.
3). A worker is chopping a forest on a grass hill. The next tile, also a grass hill, has iron but no mine. The city will finish a Factory in two turns.
4). A worker is building a windmill on a desert hill which already has a mine. The city is size 13, has wheat, cow, 5 food surplus, and two undeveloped riverside grass tiles. It is making 20 hammers at present.
5). Another worker building a windmill over an existing mine. There are two undeveloped tiles available: the city is size 10, making only 12 hammers (while the mine exists).
6). Another worker building a workshop over a farm; there are three undeveloped grassland tiles available, one with a forest - and could have a lumbermill.

On the other hand, one of my rivals (Saladin) has eleven cities visible to me and, apparently, only two workers, who are busy road-making. It's going to be slow going for my armies through all his jungle and forest.
 
1). the other two building another Workshop on a tile which already has a Farm.

So just set your options so that Workers don't replace existing improvements and quit whining. You're better off not using worker automation anyway. All these complaints are your damn fault for being lazy.
 
So just set your options so that Workers don't replace existing improvements and quit whining. You're better off not using worker automation anyway. All these complaints are your damn fault for being lazy.

Hihi... I wonder how he'll feel when he realizes... :mischief:
 
Bushface said:
Currently I have three vassals; between them, I can see the following examples of less-than-optimal behaviour :
So just set your options so that Workers don't replace existing improvements and quit whining. You're better off not using worker automation anyway. All these complaints are your damn fault for being lazy.

I think he's talking about his vassals' workers, not his own.

And why were you so aggressive in your reply? The OP is just complaining about the AI limitations in this game, nothing more, nothing less. Oh, and I agree with him on that subject.

Quit trolling. :rolleyes:
 
I'm just wondering what the super-secret code is to turn off worker automation for my vassals. ;)

Bh
 
So just set your options so that Workers don't replace existing improvements and quit whining. You're better off not using worker automation anyway. All these complaints are your damn fault for being lazy.

[sarcasm]"Good"[/sarcasm] Sir, I feel that you are the one who is lazy for not taking the three seconds out of your day to read the very second sentence of the topic creator's initial post.

I carry no ill weight on my shoulders for having poked fun at people who I feel are rude or aggressive - this being such a circumstance. Although I have not read your other 2934 posts, Willem, I do hope that they can be viewed in a different light.
 
Bhruic, if you can't find it then I bet that (unsurprisingly) there is no such code. But clearly the existing worker-automation algorithm is capable of much improvement.
Willem, your critical-response algorithm is also capable of much improvement, and apparently so are your reading skills.
(Turns off flamethrower and exits.)
 
. . . Absolute Idiocy, at least sometimes in respect of worker activities. Currently I have three vassals; between them, I can see the following examples of less-than-optimal behaviour :-
1). 4 workers in a fat cross; one building a Farm, one a Workshop, the other two building another Workshop on a tile which already has a Farm. There are two undeveloped grass tiles in the same fat cross.
.

Actually there Could be a reason for this..since Farms Also chain Irrigation, maybe the Farm being built was determined to be better for chaining Irrigation than the one being replaced by a Workshop

Otherwise it suggests a random element in determining what goes best on a tile... a bad idea

.
2). A worker building a windmill on a grass hill. All other tiles in that fat cross are undeveloped: there is copper on one grass tile, and several riverside grass tiles.

Only possibility I can think of for this is that I believe workers don't interrupt a task to start a new one, perhaps the windmill was started on before the copper was in cultural borders?
The AI does seem to overvalue windmills


.
3). A worker is chopping a forest on a grass hill. The next tile, also a grass hill, has iron but no mine. The city will finish a Factory in two turns.

Possibly the same issue as above... although less likely


.
4). A worker is building a windmill on a desert hill which already has a mine. The city is size 13, has wheat, cow, 5 food surplus, and two undeveloped riverside grass tiles. It is making 20 hammers at present.

The AI unfortunately looks at the new yield more than the change in yield from a tile improvement (when it really should ONLY look at the tile improvement) [Last I heard]

although
Mine->Windmill
v.
Add Farm

both (pre-bio) add 1 food , but the Windmill adds 1 commerce and removes ~2 hammers

so if Food is weighted highly and the Commerce is weighted sufficiently more than the Hammers for that city (and it wants food) then there could be a reason, probably not though (food shouldn't be weighted that high)

.
5). Another worker building a windmill over an existing mine. There are two undeveloped tiles available: the city is size 10, making only 12 hammers (while the mine exists).
Same as 4


.
6). Another worker building a workshop over a farm; there are three undeveloped grassland tiles available, one with a forest - and could have a lumbermill.


probably same as 4
 
The worst thing I've seen is that I got a vassal once and he REFUSED to build another city. Eventually I saw a settler in a island above me and he kept it there for OVER 200 TURNS before building a city.
 
I think he's talking about his vassals' workers, not his own.

Ooops, my bad. Sorry.

And why were you so aggressive in your reply?

Because I've seen too much complaining about trivial things in the game, my patience gets a bit worn when I think I see someone whining about a problem that doesn't even exist. But I guess in this case the complaint was justified.
 
I think AI has always been the weak point for FIRAXIS ? I mean, why does higher level has to be handicap for human player and not the other way round?

Frankly, if the AI is so good that it is almost unbeatable at the highest level (as is with a good chess program), then no one will whine (ok, perhaps they will whine that why the AI is so strong :p ). And the game will give you handicap at the lower level (like the AI is less aggrassive, expand slower, or pretty much do what it does now ;) ). I believe a good game engine should be able to offer a strong AI that get tone down on lower level and not what Civ has always offered, same AI level but handicap in terms of materials and unit, but alas. :(
 
.

1).Actually there Could be a reason for this..since Farms Also chain Irrigation, maybe the Farm being built was determined to be better for chaining Irrigation than the one being replaced by a Workshop

Otherwise it suggests a random element in determining what goes best on a tile... a bad idea

2).Only possibility I can think of for this is that I believe workers don't interrupt a task to start a new one, perhaps the windmill was started on before the copper was in cultural borders?
The AI does seem to overvalue windmills

3).Possibly the same issue as above... although less likely

4).The AI unfortunately looks at the new yield more than the change in yield from a tile improvement (when it really should ONLY look at the tile improvement) [Last I heard]

although
Mine->Windmill
v.
Add Farm

both (pre-bio) add 1 food , but the Windmill adds 1 commerce and removes ~2 hammers

so if Food is weighted highly and the Commerce is weighted sufficiently more than the Hammers for that city (and it wants food) then there could be a reason, probably not though (food shouldn't be weighted that high)

5).Same as 4

6).probably same as 4

More details.
1). No, not chaining. Tile at the edge of fat cross, desert beyond.
2). Yes, possible. Windmill almost finished: don't know when city expanded to include the copper.
3). Not possible; city at 60% cultural defence, so well beyond fat cross.
4). So automated workers go for changing existing improvements rather than developing unimproved tiles, do they ? Yet another reason why I never set them automated, and a suggestion that the automation algorithm is far from perfect.
5). and 6). I agree - same as 4).
 
3 might be caused if the Windmill was started on before iron working was discovered Or by an extreme version of favoring food (since workers are either on
1. Develop a city when they aim to maximize yield of the fat cross tiles based on that city's priority
OR
2. Trade network (connect cities, develop resources)

Workers switch between those, but only if the overall goal (enough development in that city) is done
and I think small cities focus on food significantly.


As for #4-6 (its not that they favor already developed tiles... its that the tile that is targeted is the one that will produce the best after improvement... regardless of how well it is producing now)

Number 1... unless there is some random number I can't see why (unless some of those tiles are shared by different cities)
 
3 might be caused if the Windmill was started on before iron working was discovered Or by an extreme version of favoring food (since workers are either on

As far as I read, the city was producing a factory. Since Iw is pretty much ealrier than Assembly Line, I cannot agree on this point.

Oh, and Windmills req machinery that reqs Metal casting that reqs either Iw or Bw, so getting mills before sucha n important tech as Iw is kinda bunk :D
 
^ it can happen (machinery needs BW not IW) but a Factory .. no that would need IW

for 3 I'm not sure why, unless the city REALLY wanted food, and only had hills/iron available.
 
Krikkitone - in #3 the city has adequate food and no iron.

It seems that automated worker preferences are :-
A). Changing a Farm to a Workshop is better than building the Workshop on undeveloped grassland.
B). Chopping a forest is better than building a Mine on an undeveloped hill, even when there is no urgent need of production.
C). Changing a Mine to a Windmill is better than developing a riverside grass tile, even when the city has plenty of food and there are undeveloped hills available.

I still think these decisions are ludicrous. However, if the present system keeps my opponents, as well as my vassals, from developing their lands optimally, who am I to complain ?
 
Well
A) I understand... it doesn't make sense but I understand it (it seems the worker prefers to work the tile its on as well)

B) Admittedly makes no sense Mine+Ironhill > mine+blank hill
unless it was cutting the forest to build a Windmill

C) Well OK that one i don't understand

By and large it seems the problem is how Windmill happy the AI is (as well as going for the 'best' spot to build in rather than the best Change to make)
 
Back
Top Bottom