How effective is the Automated Build Improvement?
- Thread starter Hornet85
- Start date
Another important thing is that it doesn't micro the workers to max effeciency by saving worker movement. But of course city specilization and prioritizing worked titles instead of unworked titles is probably the main reason it is bad.
Big Boss Man
Chieftain
It really only comes in handy after railroads - when your done laying down priority tiles i.e. railroads on mines and forests ect. and all thats left is laying rail ... Set to auto trade network after you have linked all your cities with rails thats when i do it ( at least some of them anyways ) .
Stephanus
Chieftain
IMHO the automated worker only does random improvements to each tile (besides laying a road or railroad that is).
I prefer giving orders to the workers myself, although this is taking a lot of time compared to automating them.
I prefer giving orders to the workers myself, although this is taking a lot of time compared to automating them.
Carabodes
Timid Viking Warlady
I never automate my workers. Automated workers do crazy things, like:
One worker tears down a cottage to build a farm, while simultaniously on the neighbouring tile another worker tears down a farm to build a cottage.
When not at work, they hide out in cities. If an improvement then is needed in another city some distance away, the sleeping workers do not pick up on that and continue
Or, if they do pick up on a tile in need of work, all the workers rush to the same tile whilst other tiles in need of work will not be touched.
In addition, all automated workers seem to spend a lot of time building forts all over the place.And they roughlessly chop down every forest, leaving no opportunity to build lumbermills or forest reserves.
And if you mark for "Automated workers leave forests" and "Automated workers leave old imrovements" they will not improve any jungle or forest tile until replaceble parts, then they build lumbermills everywhere (in forests) and when forest reserves becomes available, every jungle square becomes a forest reserve. They simply don't care about the terrain under the forest/jungle.
As for leaving old improvements, if they have made a mine on a hill, that mine will never become a windmill, and if a late resource (like coal) appears where there is a windmill, they'll never build a mine.
They know that something is needed, though, so they pile up on the spot and "hang out", waiting for something to happen.
And all of them just stay there, idle on the spot, whilst other tiles do not get worked due to lack of available workers! 
And then the cities will start begging you to build more workers, since you obviously havent got enough of active workers.
And if you put them on automation for building trade network, they do start off by building a road from one spot to another, but when passing a resource, they get "sidetracked", leave the roadbuilding and start improving the resource. And then they spot another resource and .... ..... ..... What happened to the roadbuilding?
Automated workers =
That automated workers are inefficient, bad, whatever, has been said a few times, so let's look at the case a bit deeper.
Workers build tile improvements, including roads. Tile improvements are built both within and out of the fat crosses of cities, roads even outside cultural boundaries.
Roads are commonly needed to allow for fast passage of units, to connect cities to trade network, and to connect resources to trade network. The latter two can be done in optimal manner assuming that's all that needs to be done, and I would expect AI controlled workers (ie. automated workers) to handle that task fairly even if not well.
Tile improvements are built to access resources and to gain benefit from working a tile. When outside fat crosses of cities, only resources are interesting and with BTS fort works just as well as resource-specific improvement. Outside fat crosses workers should handle that well even when automated.
But when we consider improvement meant to be worked by a city, we have to go somewhat deeper in our thinking. Down to "what is the purpose of this city?" which can be answered in many ways. For some cities you want to prioritize mining the hills for max hammers, for others you want to farm the riversides, yet another city type could want cottages. Or maybe you want watermills riverside?
Want to keep forests for health, maybe lumbermill them later? Or if you keep lots, maybe forest preserves and national park? Or you want to chop them for production now as you're building a wonder? Or in case your workers have some idle time (shouldn't happen, so let's call it "no immediate critical needs" -time) they can prechop so that you can chop fast when you need the hammers soon for another wonder, or to get army up, or whatever?
The AI can't know what you want out of that city. So it can't know whether to prechop or chop now, where to road towards, which improvements to build.. Thus, automated workers build an improvement based on terrain. "Hmm, looks like we have some riverside grassland here, and that's usually good for cottages, so let's do that". Err.. NO! Now they chop the precious forest I was saving for Oracle
Workers build tile improvements, including roads. Tile improvements are built both within and out of the fat crosses of cities, roads even outside cultural boundaries.
Roads are commonly needed to allow for fast passage of units, to connect cities to trade network, and to connect resources to trade network. The latter two can be done in optimal manner assuming that's all that needs to be done, and I would expect AI controlled workers (ie. automated workers) to handle that task fairly even if not well.
Tile improvements are built to access resources and to gain benefit from working a tile. When outside fat crosses of cities, only resources are interesting and with BTS fort works just as well as resource-specific improvement. Outside fat crosses workers should handle that well even when automated.
But when we consider improvement meant to be worked by a city, we have to go somewhat deeper in our thinking. Down to "what is the purpose of this city?" which can be answered in many ways. For some cities you want to prioritize mining the hills for max hammers, for others you want to farm the riversides, yet another city type could want cottages. Or maybe you want watermills riverside?
Want to keep forests for health, maybe lumbermill them later? Or if you keep lots, maybe forest preserves and national park? Or you want to chop them for production now as you're building a wonder? Or in case your workers have some idle time (shouldn't happen, so let's call it "no immediate critical needs" -time) they can prechop so that you can chop fast when you need the hammers soon for another wonder, or to get army up, or whatever?
The AI can't know what you want out of that city. So it can't know whether to prechop or chop now, where to road towards, which improvements to build.. Thus, automated workers build an improvement based on terrain. "Hmm, looks like we have some riverside grassland here, and that's usually good for cottages, so let's do that". Err.. NO! Now they chop the precious forest I was saving for Oracle
Powerslave
Prince
I was wondering, will a worker build the most optimum improvement for each tiles if I set them to automatically build improvement? Or should I manage each improvement myself?
Thank you... I'm just a beginner to this game
You know what? Automating your workers is perfectly okay. Go ahead and do it, if you want. They just aren't that bad at the job any more (they used to be horrible).
Just make sure that you set "workers leave old improvements", so that you don't get one worker making a cottage, and then have a second worker destroy the farm to make a cottage.
I can win on Monarch by automating my workers and setting the cities to use the governor. I might even be able to win on Emperor with automated workers, but I haven't tried that yet.
I would, however, suggest that you control the workers yourself, if you want to become a better player. I can (and do) take control of my workers when I see them do something bone-headed.
And you don't need to specialize your cities, either. Go ahead if you want, but it's not necessary until the higher difficulty levels. You can just cottage everything and cruise to victory on Monarch, even with non-Financial leaders.
The only time I ever use automatation on workers is after they've done all of the significant work and can be set to "automate trade routes" or whatever it is called, where they will railroad everything for you.
If you automate them in general, I recommend not only "leave old improvements" but instruct them to leave forests alone, as you will probably want to direct your own chopping operations to enhance production in cities.
If you automate them in general, I recommend not only "leave old improvements" but instruct them to leave forests alone, as you will probably want to direct your own chopping operations to enhance production in cities.
"I can win on Monarch by automating my workers and setting the cities to use the governor."
I find this very hard to believe. The city governors are actually quite horrible and worse than before BTS. That's impressive if you are giving yourself that major handicap and still winning a game on Monarch level! Are you just using the city governor for growth or also production decisions? and do you literally just leave a city automated for the whole game? how do you manage specialists, because the governors are actually broken when it comes to this part - they won't assign the right kind of specialists at all for what you need.
using the city governors is basically just letting your civ's inner mechanics be run by another AI. If you do that, all that's left for you to decide is where to move your units, what techs to research next, and what civics and trades to make.
I find this very hard to believe. The city governors are actually quite horrible and worse than before BTS. That's impressive if you are giving yourself that major handicap and still winning a game on Monarch level! Are you just using the city governor for growth or also production decisions? and do you literally just leave a city automated for the whole game? how do you manage specialists, because the governors are actually broken when it comes to this part - they won't assign the right kind of specialists at all for what you need.
using the city governors is basically just letting your civ's inner mechanics be run by another AI. If you do that, all that's left for you to decide is where to move your units, what techs to research next, and what civics and trades to make.
KrikkitTwo
Immortal
- Joined
- Apr 3, 2004
- Messages
- 12,418
Automate workers has the following problems
1. not good at specializing (although apparently they 'listen' to your city focus now)
2. wasting worker movement
3. tearing down things for those that are 'better'
4. workers are too risky (putting themselves where they can be captured)
You can use automated workers well if
1. You have "Leave old improvements" + "Leave forests"
2. You are always grabbing some workers to do important things*
* specialize a city, get an irrigation chain optimized, put watermills in the optimum places, put forest preserves around your National Park city, chop forests that need chopping
They help take care of a large empire, so that your generic cities and extra resources get developed.
It would give you a benefit to plan every move of every worker out... but there are strongly diminishing returns
1. not good at specializing (although apparently they 'listen' to your city focus now)
2. wasting worker movement
3. tearing down things for those that are 'better'
4. workers are too risky (putting themselves where they can be captured)
You can use automated workers well if
1. You have "Leave old improvements" + "Leave forests"
2. You are always grabbing some workers to do important things*
* specialize a city, get an irrigation chain optimized, put watermills in the optimum places, put forest preserves around your National Park city, chop forests that need chopping
They help take care of a large empire, so that your generic cities and extra resources get developed.
It would give you a benefit to plan every move of every worker out... but there are strongly diminishing returns
troytheface
Deity
- Joined
- Mar 26, 2002
- Messages
- 3,262
I automate everything but Kirkktones two points above are worthy of note.
(Leave old improve/leave forests sometimes grab a worker to rush chop a wonder- and i would add - to road your army to its destination) But you have to check once in a while so when they build a fort on top of say like iron you can go back and mine it.)
The strength of full automation (in this case missionaries and corporations)too is that missionaries succeed more often in converting your cities to your religion. (Less "Failed to convert city" in red messages)
Also full auto is one of those self imposed challenges- in so far as immersion i say it adds realism (workers not always on task...)
in short, reasons I automate ....
Frees focus to spying and military
Avoids fixating on building a farm when i should be working that horse...(auto preference -choosing a better option)
Better conversion ratio (missionaries)
Surprise builds that end up favorable (a fort serving as a port ...and by god i can use it ....)
More modern play style instead of stodgy micro managing nightmare horror play
(Lincoln/Standard/Random Map/Prince)
(Leave old improve/leave forests sometimes grab a worker to rush chop a wonder- and i would add - to road your army to its destination) But you have to check once in a while so when they build a fort on top of say like iron you can go back and mine it.)
The strength of full automation (in this case missionaries and corporations)too is that missionaries succeed more often in converting your cities to your religion. (Less "Failed to convert city" in red messages)
Also full auto is one of those self imposed challenges- in so far as immersion i say it adds realism (workers not always on task...)
in short, reasons I automate ....
Frees focus to spying and military
Avoids fixating on building a farm when i should be working that horse...(auto preference -choosing a better option)
Better conversion ratio (missionaries)
Surprise builds that end up favorable (a fort serving as a port ...and by god i can use it ....)
More modern play style instead of stodgy micro managing nightmare horror play
(Lincoln/Standard/Random Map/Prince)
KrikkitTwo
Immortal
- Joined
- Apr 3, 2004
- Messages
- 12,418
One point. forts are better than Mines if they are Outside your city radius
troytheface
Deity
- Joined
- Mar 26, 2002
- Messages
- 3,262
well then, no wonder they build forts. I still get horrifed to see forts on oil and marble and will rebuild them to refinerys and quarries...seems like you would at least get the resource bonus - when i click the tile improvement to see if a fort provides anything it shows...fort stuff. when i rebuild it shows the resource.
Which is another reason to automate. Good for a player that refuses to read stuff about tile improvements and city radius' and the like.
Which is another reason to automate. Good for a player that refuses to read stuff about tile improvements and city radius' and the like.
