Worker Suggestions
- Thread starter Djinn
- Start date
Mostly on luxuries you should follow them and constructing roads you should also follow them. But sometimes you have to think what you really want. Like when you have a pyppet they will adcvise you to build trading posts only but you can build farms to increase growth or mines to increase production. First get all the luxuries and other tile improvements by yourself and later when you don't have much tiles left you can put them automatic for a short time. 
sufficiency
King
- Joined
- Feb 6, 2013
- Messages
- 682
Post Turn-150ish I usually start to rely a lot on the automatic stuff, esp. if I have a wide empire.
The automatic improvement algorithm is horrible. It cannot optimize road placement by taking building an "Y" connection to three cities into account. It will shop too many jungles instead of putting trading posts on them. It cannot differentiate between own cities you want to grow as much as possible for specialists and cities that you want too keep small and just bring in gold.
Especially in the beginning you need to consider carefully what tiles to improve. First priority is to make sure city (especially capital) has enough farms. Second priority is some production. If you want to have a good game micromanagement is essential and in my opinion good also for the immersion. Turning on automatic improvement is not only inefficient but also boring.
The only automatic movement I use is for exploration once I think I've the biggest other civs. This algorithm is fine for the random movement required to find the last natural wonders. I also find the shuttling back and forth of the map view that three or four ships in all parts of the world makes for disturbs concentration.
Especially in the beginning you need to consider carefully what tiles to improve. First priority is to make sure city (especially capital) has enough farms. Second priority is some production. If you want to have a good game micromanagement is essential and in my opinion good also for the immersion. Turning on automatic improvement is not only inefficient but also boring.
The only automatic movement I use is for exploration once I think I've the biggest other civs. This algorithm is fine for the random movement required to find the last natural wonders. I also find the shuttling back and forth of the map view that three or four ships in all parts of the world makes for disturbs concentration.
Situation dependent:
If there is a luxury, bonus, or strategic resource present, yes that's reliable.
Road placement: If playing Inca you may want to ignore it and go for a longer route that's 100% on hills. And playing the Irq. there may be a different tile that saves more roads (using forest)
Other than that it's decent unless there are 3rd city considerations.
Other improvements: Ignore especially if you empire is temporarily unhappy or in a GA. (Being in a GA will cause it to recommend trade posts every where; being unhappy will cause it NEVER to suggest farms.)
I'd hate to see what an automatic worker would do, especially if you let it override existing improvements.
Biologist
Emperor
I typically ignore the improvement suggestions. Generally if I happen to agree with the suggestions then I would have placed that improvement there anyway. The only function I will occasionally automate is building short roads between cities in relatively safe locations. Everything else I tend to micromanage.
It might be amusing to play a game where one only follows the tile improvement advice for worker actions, only settles cities in recommended locations, and only makes the research/construction choices suggested by the advisors. I think this has been done before, but not for a while.
It might be amusing to play a game where one only follows the tile improvement advice for worker actions, only settles cities in recommended locations, and only makes the research/construction choices suggested by the advisors. I think this has been done before, but not for a while.
No need for that, any time after you have visibility into AI lands; just look at their portion of the map.
AI workers depend on general and broadly recommended rules and for low difficulty this is fine and for higher difficulties the worker advice is worth considering
But the AI cannot read your mind, you should have a plan for how to develop your cities. Simple example is banana in jungle - what is right improvement? It depends on what your plan for the city is. Science or growth? If you are going ics it is probably science (no improvement) and if you are going tallish it is probably growth (plantation). The AI in civ 5 has of current patch and with reasonably available hardware no way of reading your mind. Until that happens the better option is probably to do the micromanagement
. yes, i hate mm of workers too but it pays off quickly
But the AI cannot read your mind, you should have a plan for how to develop your cities. Simple example is banana in jungle - what is right improvement? It depends on what your plan for the city is. Science or growth? If you are going ics it is probably science (no improvement) and if you are going tallish it is probably growth (plantation). The AI in civ 5 has of current patch and with reasonably available hardware no way of reading your mind. Until that happens the better option is probably to do the micromanagement
. yes, i hate mm of workers too but it pays off quicklyIn this case, the choice is more do you improve it or not; your not allowed to put anything except a plantation on bananas, which will clear the jungle.
The same goes for every other resource improvement in Civ V. *
* With the minor exception of strategic resources you don't have the tech to see yet.
Yes, bananas are to improve or not to improve, but was used as a simple example compared to the regular TP vs farm vs mine choices. But let's talk about resources and eventually bananas
The automated workers are useful in maybe 80-90% cases and I have no problem recommending them for the beginning player so they can focus on learning the many other aspects of civ5. For people playing on harder difficulties these workers do too many "wrong" things because we typically have a long-term strategy which the AI has no way of knowing. The auto workers are really fond of food and seems to favour improving those tiles in most cases, but does not seem to take into account your strategy and other gameplay aspects like happiness control or faith/pantheons.
Automated workers are also fond of exploiting resources, and generally that is ok, especially where things can be reverted. Jungles and forests cannot be unchopped (with an unmodded civ5) so if a worker screws up you "lose" that tile for the rest of the game. For certain strategies it is a significant boost to keep jungles' science boost, but the auto workers will see a banana, become hungry and will step over to add a plantation for more yellow curved food. And you notice too late, naturally, when you get the message "14 hammers added to My City".
I want to use auto workers because it is not fun micro manage them after turn 100-150 when the critical resources have been handled. It is mainly a UI issue I think - the two worker options and preferences are too few. Maybe a screen to define basic rules for workers could make them a more viable option for turns 150+. Some of the things that annoy me are especially the destructive operations, how workers handle road quests and priorities hammer/food/money/science/luxuries, ... Most of those should be possible to handle in a single screen combined with some rules about what areas a certain worker should focus on.
I have not tested auto workers with hiawatha or the celts, do they take into account forest preferences?
Aaron90495
King
Yes, bananas are to improve or not to improve, but was used as a simple example compared to the regular TP vs farm vs mine choices. But let's talk about resources and eventually bananas
The automated workers are useful in maybe 80-90% cases and I have no problem recommending them for the beginning player so they can focus on learning the many other aspects of civ5. For people playing on harder difficulties these workers do too many "wrong" things because we typically have a long-term strategy which the AI has no way of knowing. The auto workers are really fond of food and seems to favour improving those tiles in most cases, but does not seem to take into account your strategy and other gameplay aspects like happiness control or faith/pantheons.
Automated workers are also fond of exploiting resources, and generally that is ok, especially where things can be reverted. Jungles and forests cannot be unchopped (with an unmodded civ5) so if a worker screws up you "lose" that tile for the rest of the game. For certain strategies it is a significant boost to keep jungles' science boost, but the auto workers will see a banana, become hungry and will step over to add a plantation for more yellow curved food. And you notice too late, naturally, when you get the message "14 hammers added to My City".
I want to use auto workers because it is not fun micro manage them after turn 100-150 when the critical resources have been handled. It is mainly a UI issue I think - the two worker options and preferences are too few. Maybe a screen to define basic rules for workers could make them a more viable option for turns 150+. Some of the things that annoy me are especially the destructive operations, how workers handle road quests and priorities hammer/food/money/science/luxuries, ... Most of those should be possible to handle in a single screen combined with some rules about what areas a certain worker should focus on.
I have not tested auto workers with hiawatha or the celts, do they take into account forest preferences?
With regard to Hiawatha or the Celts, no, they don't take into account forest preferences. As well, workers will ALWAYS build a trading post (if possible) when you're in a Golden Age, as they think the yield is higher permanently. I personally refuse to automate workers, even in the end-game; there's always stuff they mess up, and if I don't need them much, I'll just delete them. Maintenance is really high late in the game, and if they're not improving crucial tiles that I'll use, they're not worth the GPT.
compwiz1202
King
I agree, I wish they would expand on the focus function to also make auto workers build things to coincide with your city focus. I always manually improve my capital. Sometimes I will use auto for wider games once my capital is established if I have roads set up so I have a better chance of getting my worker back to safety if barbs show.
Zezima
Professional Troll
I never use auto. The workers make a disaster of everything.
The only reason I would ever see it being useful is if you have a massive sprawling empire, at which point automating the workers partially defeats the purpose of having the large empire in the first place.
Also minor nitpick, jungles don't give production when chopped.
The only reason I would ever see it being useful is if you have a massive sprawling empire, at which point automating the workers partially defeats the purpose of having the large empire in the first place.
Also minor nitpick, jungles don't give production when chopped.
Smokeybear
Emperor
Amen, brutha. Automation is just asking your workers to shoot you in the foot.
