Terrain Improvement?

[Skanky Burns]

How would you like the terrain around cities to be improved?

Civ3 style - Workers working the land. Physical units:
You build workers, send them out to a tile near (or far) from your city and then order them to build improvements. Micromanagement hell towards the end-game.

CtP-style Public Works:
Could someone who has played CtP give a description of how this works? Thanks.

Terrain automatically improved over time:
The terrain around your cities improves automatically over time. Roads are built between cities that trade together (including towards foreign cities you trade with). You, of course, have no control over what gets built where or when, apart from perhaps setting "policy".

[Fosse]

I voted Public Works. Perhaps if the "Over Time" approach were more fleshed out I would have gone that way. But The micromanagement disaster that is the Worker system has simply got to go. The AI can't handle it, and the humans shouldn't have to.

[wrylachlan]

I voted other because a though the worker system is a pain, public works takes away the possibility of capturing workers, which I like. What I'd like to see, and I've said this before, is workers like in Civ3 with a Public Works Manager on top of that system. So you could control workers individually, or, if they don't have anything to do, they would automatically "check-in" with the Public Works Manager to see if there's anything that needs to get done.

As a side note, a number of different variations on workers vs. public works were put forward in various other threads. I'd like to see a poll that lists all of the different options people have come up with. I have a feeling that this poll will skew towards Public Works, even though a lot of people who vote for it might vote for a different hybrid system given the choice.

[Skanky Burns]

If you could post some links to the other improvement systems or at least descriptions of them, we can restart this poll with a much better range of poll options.

I will be the first to admit I don't know much about improvement systems other than workers, but this is an issue that does need to be discussed.

[Ozymandous]

It's been a few years since I have played CtP, but here's what I remember about "public works."

A Civ has to dedicate a percentage of it's income per turn to building a "fund" for Public works. Improving each tile costs X amount. Let's say you want to put a "level 1" fam over grassland. It costs 5 public works for the one tile. When you have accumulated 5 or more "credits" into the public works fund you then select to upgrade that tile and it's done the beginning of the next turn.

Each improvement, from roads, to irrigation to mines to nets (for fishing) costs some amount of public works "credit". Some improvements can't be placed over certain terrain types. There are/were three different levels of public works, and each improvement above level 1 comes when certain technologies are developed.

The best thing about this system was that the more you allocated to the public works fund, or the more your economy did with a lower percentage going into the fund, the more you could upgrade each turn. Initially you might upgrade a tile every few turns, but after awhile you could upgrade a lot of tiles at once. Plunk down a new city and build the roads to it and upgrade a few tiles in that first turn. No having to coordinate workers, no having to "plan ahead" as much as in Civ3 now. The AI seemed to use this system much, MUCH better than now.

I hope something like this is done in Civ4, since it take the "worker hell" micromanagement in mid-late games nonexistant.

I hope this helped explain the system for those who have never seen it.

[Kuciwalker]

I want the things to be represented by a unit at some level.

[asleepathewheel]

civ isn't civ without individual worker type units

[Ecthy]

I like the Master of Magic approach

[Fosse]

Originally posted by asleepathewheel
civ isn't civ without individual worker type units

Or really dumb AI moves. Let's face it, every Civ game ever has had really dumb AI moves in it. Thus, not having them would be Not Civ.




If this is the case, then I move that Firaxis make NotCiv 4, a game that removes dumb AI moves and also individual worker units.

[hexagonian]

...Some kind of hybrid system, as pointed out by wrylachlan - it would require an overhaul of the existing economic system that may have to multiply normal shield output of a tile by (2x), or some multiplier to be determined by the developers...

You have a national and/or a local city pool of shields that can be used for your improvements. Workers are not required if you want to build within a city radius either.

So if you want to build a new mine in a city, you will require 8 shields to do so. The construction of the mine without an assigned worker will take 8 turns to construct.

A worker can be created to reduce construction time by a turn for each worker - not by 1/2 time as it is currently in civ3. If you want to build outside the radius, at least one worker is required to do so. You also make workers expensive.

The end result is that workers are needed, and are valuable in the early game, especially in roading to a trade good and pre-establishing a site to be settled - but their increased cost, their reduced effectiveness, and the aspect that you will still be able to build tile improvements without them, will actually reduce the number of workers in the game.

[wrylachlan]

Originally posted by hexagonian
...Some kind of hybrid system, as pointed out by wrylachlan - it would require an overhaul of the existing economic system that may have to multiply normal shield output of a tile by (2x), or some multiplier to be determined by the developers...

You have a national and/or a local city pool of shields that can be used for your improvements. Workers are not required if you want to build within a city radius either.

So if you want to build a new mine in a city, you will require 8 shields to do so. The construction of the mine without an assigned worker will take 8 turns to construct.

A worker can be created to reduce construction time by a turn for each worker - not by 1/2 time as it is currently in civ3. If you want to build outside the radius, at least one worker is required to do so. You also make workers expensive.

The end result is that workers are needed, and are valuable in the early game, especially in roading to a trade good and pre-establishing a site to be settled - but their increased cost, their reduced effectiveness, and the aspect that you will still be able to build tile improvements without them, will actually reduce the number of workers in the game.

Just out of curiousity, what is the advantage of using two different systems vs 1? I don't envision a Public Works director as being seperate from the workers, but rather an overlay on top of the worker system. All workers do all the work just like in Civ3. The difference is in the way you give workers orders. Right now you have to give them individual orders. Instead I'm suggesting that the worker units could get their orders en masse from a public works director. Thus you could EITHER order a worker to mine a certain tile, OR you could tell the Public Works Manager that you want a certain tile mined, and the Public Works Manager would do the tedious work of finding which worker is closest and assigning him to the task.

The Public Works Manager could also give you some high level feedback in terms of "The Public Works you have requested will take a total of 29 turns to complete with the current work force. If you want to decrease that time, please build more workers."

[Fosse]

I like the idea of the Public Works Manager wrylachlan is putting forward as a reasonable compromise between two systems. It will allow the ultra micromanagement nuts to hand move every single worker to their little hearts' (whose beats' control has no doubt been ceded by their unconscious to their conscious brain) content, while allowing those of us who find this tedious to simple give orders.

I would rather push it one step further into the strictly Public Works systems, where you fund public works from your economy and simply issue orders, no workers involved. But this is a happy compromise that I would be glad to have.

[hexagonian]

Originally posted by wrylachlan
Just out of curiousity, what is the advantage of using two different systems vs 1? I don't envision a Public Works director as being seperate from the workers, but rather an overlay on top of the worker system. All workers do all the work just like in Civ3. The difference is in the way you give workers orders. Right now you have to give them individual orders. Instead I'm suggesting that the worker units could get their orders en masse from a public works director. Thus you could EITHER order a worker to mine a certain tile, OR you could tell the Public Works Manager that you want a certain tile mined, and the Public Works Manager would do the tedious work of finding which worker is closest and assigning him to the task.

The merging of the two different systems would require the use of shields (a cost factor over and above the initial cost of a worker, so you will have to plan this cost out), and it would allow for the capture of workers because they are an actual unit. Best of all, it would allow for the construction of most types of Tile improvements without the tedious efffort of issuing orders to an army of Workers, because they become less important. You still have them but they augment your efforts rather than forcing a player to solely depend on their efforts to get the job done. So you get the best of both worlds.

The way I see it, you proposal is not much different than the automation command feature currently in place - a feature that some players still complain about because it ends up sending units willy-nilly around your empire. You may order a unit via the Director to build a tile improvement and it may take that unit 4 turns to get there to start it - meanwhile another worker is finishing up another improvement and will be closer than the first worker to that tile on the next turn - and so you will need to issue orders to that unit on the following turn. You then run into the problem of having to reissue an order based on efficiency.

[Fosse]

wrylachlan's proposal is quite a bit different than the current automation implementation, I think.

Instead of saying to the workers, "Do as you will! Have fun!" or giving them only the vaguest commands, "Don't mine my farms, but otherwise go hogwild!" wrylachlan is saying that precise orders are given and carried out.

So if I decide I want to build a farm next to London, I just right-click on that tile, and select "Order Farm." My job is done. Now the Public Works manager will find the nearest worker who is not busy and order him to move to that tile and build that farm.

The AI for the Manager would need to be done well enough to let the Manager calculate whether it is more efficient to bring an idle worker from the other side of the empire, or simply wait for a nearby one to finish his next task. The Manager would do all of the ganging, and figure out the annoying details for us.

If we wish, we can manually move the workers around and do things as we currently do, always having the option to turn them over the the PWM.


This is MUCH different, and better, than the current automation.

[wrylachlan]

Originally posted by Fosse
wrylachlan's proposal is quite a bit different than the current automation implementation, I think.
{snip}
The AI for the Manager would need to be done well enough to let the Manager calculate whether it is more efficient to bring an idle worker from the other side of the empire, or simply wait for a nearby one to finish his next task. The Manager would do all of the ganging, and figure out the annoying details for us.

If we wish, we can manually move the workers around and do things as we currently do, always having the option to turn them over the the PWM.

This is MUCH different, and better, than the current automation.

Thanks, you beat me to it. To build on that in order to answer Hex, the AI really is the sticking point. The problems you point out about workers crossing the continent, vs. the closer worker finishing his job, etc. etc. is a problem not of the basic idea, but a possible problem of the implementation of the AI.

On a side note, if you are attached to the idea of improvements costing shields, one possibility is to simply assign a shield upkeep cost for every turn that a worker is working a tile. The upkeep is drawn from the nearest city. I'm personally not sure whether this is really necessary, but it is a possibility that is not anathema to the suggestion I've already put on the table.

[hexagonian]

Well, the main sticking point for me with workers has always been the endless moving of units and then issuing the build commands for them once they got to the tile. With PW, you did not have to deal with moving units - just click an improvement on a tile and it would be constructed after a few turns. Very streamlined...

So if the AI could effectively manage worker movement, and all you had to do was click on a tile to select the type of improvement to build, then I'm sold on the idea. At least in SMAC, there was a multiple command option (build mine/road on a tile) This helped the micromangement problem to some degree.

Shield cost was merely the means to attach a cost to PW so it's of secondary importance.

[Brent]

Preparing Sites for Cities

I want to be able to tell workers (or public works) to prepare a new site for a city whose settler is not ready yet. Automatically build some roads or whatever in a proposed radius. You can specify a specific site, or let automation determine a good site and then another and another. Possibly use the existing Road To command and after that road is built, tell other workers to start preparing sites along that road, beginning at your current territory. Preparing city sites beforehand can lead to your opponents stealing your sites.

[Skanky Burns]

I think the system where you order improvements for the land and then have workers automatically assigned to an available task is a workable system. This allows for the hardcore player to still manually control his workers, and for the rest of us to have some control over what improvements are built (as opposed to current autoworkers) and be able to keep workers aside for projects we need done immediately such as trade roads.

Combine this with some high-level tools such as a brush (exact number of tiles it affects may vary, but a city-radius sounds good) that orders all tiles under it to have roads and/or rr built on them, irrigate all valid tiles, mine all valid tiles, etc.

With that, I actually voted in my own poll finally. Other option.

[wrylachlan]

Originally posted by Skanky Burns I think the system where you order improvements for the land and then have workers automatically assigned to an available task is a workable system. This allows for the hardcore player to still manually control his workers, and for the rest of us to have some control over what improvements are built (as opposed to current autoworkers) and be able to keep workers aside for projects we need done immediately such as trade roads.

And one of the other really cool corrollaries is that it requires zero recoding of the worker AI. Since it is possible to control the workers individually, the AI can simply reuse (and hopefully improve upon) the Civ3 codebase.

And I have one more suggestion to add -
The ability to combine workers into superworkers.

I know this sounds kind of out there, but I think that the total number of units in Civ3 gets to be a problem, and workers are no exception. I propose a Small Wonder called Guild Hall(obviously the name sucks, but that can be another discussion). Once the small wonder is built, you can combine 2 workers in a tile to form a Skilled Worker whose work output is something like 125% of what the two workers could do combined.

This would make an incentive for the player to cut down on the total amount of units in the late game. If you found there were still too many units, you could even go one step further and use the Workplace Engineering SW to allow 2 Skilled Worker to be combined into an Engineer.

[Kuciwalker]

I'll join in and say that I really like wrylachlan's proposal.

[wrylachlan]

Originally posted by skywalker
I'll join in and say that I really like wrylachlan's proposal.

You're not going to believe this, but I just looked out my window (9th Floor) and there was this huge hog with wings flying by.

[Kuciwalker]

[Pedrunn]

I voted Public Workers because is more similar to my idea.

Which is to place Tile Improvements by choosing from a menu and placing it in the map with by clicking with the mouse in the tile ypu wanted. Just like the PW system. But instead of costing PW it costs Gold.

No need to add a whole new variable (PW)!!!

[Bleyn]

I would prefer something that is at least a variation of the current system, and am slightly indifferent to the CtP PW system since it really is just the same thing at a slightly higher level of abstraction(at least as it was explained here). But I am dead set against the idea of a system where you just express preferences and things happen on their own.

While we (almost) all hate micromanagement, there is one thing that is worse than having to micromanage. That is being unable to micromanage. Especially something as critical to game flow in a Civ game as terrain improvement is.

Imagine not being able to give commands directly to the workers in Civ3, and only being able to control them through an interface like the City governer one. From my experience, no matter what emphasis you tell the governer to go with, it seems to get totally stuck on food and growth, even in cities totally surrounded by grasslands that will take forever to grow large enough to starve.

Unless a massive amount of work were done with the AI, that is probably about what we'ld end up with if Civ4 had an automatic terrain improvement system. And personally, while I wouldn't mind seeing enough work done to make worker AI worth trusting, I wouldn't want to be dependant on it.

[Trifna]

I think that CtP style with a way to give a geographical and "type of improvement" orientation could be the best.

[Brent]

Be able to assign a worker to improve an ally's terrain but still be mine. Have an automation option for a worker to find a resource to build a colony.

[Adagio]

I voted for "Terrain automatically improved over time", though with more user-control:

E.g. in the cities there should be an option where you set priority for food, shields, roads etc

[Bleyn]

Just out of curiousity, have you played MOO3 ADG? If you had, you might know what I mean when I say being unable to micromanage is worse than having to micromanage.

Specifically the planetary regions and DEAs. Sure you can manually order the construction of the DEAs themselves, but there is absolutely no way to order the construction of the enhancements to the DEAs. As a result, I find that it is almost not worth even trying to pay attention to that portion of the game. Essentially, I feel it is a matter of 'If I can't control it, I don't even want to see it. But if I can see it, I had better have some way of manually controlling it.'

The research model in MOO3 is also a good example of this. You can't actually choose what techs get researched. You can't skip a tech that you don't want. All you can do is adjust the amount going to each of the tech areas. For me, it basically became another area that I ignored. Not because I wanted to be able to ignore it, but because I had no reason not to.

That is what I am afraid automatic terrain improvement without any manual control would turn into.

[Spiffor]

I vote for CtP style public works, but with a little variation:

1. I think public works should be paid in gold rather than in a "public works reserve". The PW reserve was a useless clutter. Gold does the trick exactly as well

2. I think there should be a small food cost associated with a public work.

3. I think it should be possible to quicken a public work. One click = normal time to completion. Two clicks= twice as fast (and twice as expansive) etc.

4. A CtP style system requires a good and instinctive interface. Draw your roads like in Sim City. Use the Shift-click to cover all them connex grasslands with irrigation, just like in the Sims, when you cover a room's floor with fitted carpet. When you're deciding a PW project (before the construction actually begins), have it show in red when it's far from completion, yellow when it's near completion, green when it'll be built instantly. Etc.

5. Have some automation in the PW system. Have a PW advisor (similar to the science advisor), whom you grant X% of your income each turn, and to whom you give priorities: food, connectivity, development of a specific city, etc.

Automation and interface are very important in a CtP style PW systems, because those were the shortcomings that made the PW system a tedium rather than a progress.

[The_Aussie_Lurker]

Not sure if someone has already mentioned this possibility but, how about you have a public works budget, as in CtP, bus STILL have workers? In this system, the amount of work a worker can perform is limited by your PW budget! That way, even if you have an army of workers, you can't do much with them if you don't have the neccessary budget (for both labour and materials!)
Industrious Civs would be able to generate more PW points, and there would still be workers to capture in the game-the advantage being that you can build terrain improvements for at most half the PW cost of a normal worker.
In addition, you could allocate additional PW resources to certain workers, which would increase the speed at which they finished their work AND/OR you could allocate the PW budget more generally, so that certain Worker tasks (like mines, fortresses) were done much quicker, but only by sacrificing speed on other tasks!!
Anyway, just a few thoughts!

Yours,
The_Aussie_Lurker.