The 'size six' strategy

[La Fayette]

Most experienced players probably remember the 'size five' strategy designed by Xin Yu for civ2.
I have chosen the title of this thread in honor of Xin, our master.

I started looking at civ3 very recently (in fact a week ago) because it seems that the French version of the game is not going to be available before 2002.
I must heartily say thank you once more to another civ2 master, namely solo who, knowing that, sent me the game directly from the US by airmail.
Anyway, here is my first proposal as far as civ3 is concerned:

Winning at civ3 (just like many many other strategy games) is very much a matter of growth rate. If you grow faster than the AI, you win, ...
In the early game, the growth of your cities is related to food surplus (1 food = slow growth, 2 food = 'normal', 3 food = highly desirable). This has been very much described and stressed on this forum AFAIK.
Later on, civ2 gave us settlers (able to join the city up to size 8), food caravans and WLT*D (if Republic or Democracy).
civ3 gives us workers and settlers, able to join the city any size.
Specializing some cities in 'worker-nursing' and sending those workers where needed, and especially joining the other cities, is a nice strategy.
My proposal is to build those nursing cities UP TO SIZE 6 and let them work at that size.
There is a huge advantage doing that:
instead of building a worker once in a while (according to the food surplus in your nursing-city), your size-six-nursing- city will build one worker/turn (if your city is able to produce 10 shields/turn, food surplud at least one) or one worker every second turn (if your nursing-city provides at least 5 shields/turn and has at least a food surplus of one).
I have done that over 100 times in my current game and it works like a charm. IMO it is not a bug. It is a threshold effect due to the fact that the size of the foodbox changes at size 6 (there was the same kind of threshold effect in Xin Yu's 'size five', due to the fact that, from size 5 on, you can have specialists in your city in civ2).
Have a try, my friends. IMO it's a winner.

(La Fayette, looking at a nursing-city size 6)

[zedar]

I can understand the 1 worker per turn from a sheilds point of view, there must be something i haven't noticed with the food box though.

size 6 city builds worker and shrinks to size 5
at this stage what exactly happens to the foodbox? is it fully empty in which case you'd need alot of extra food to get your pop point back in the same turn.. or does something else happen?

At work at the moment so i cant test it.. but thinking alot about the optimal ways to build workers and settlers.. if this can somehow be applied to have a city roll between size 4 and size 6 building a settler every 2-3 turns it will be great

Zedar

[barefootbadass]

With a granary and a few good food squares(like floodplain with wheat) you could pump out settlers about as fast as you could get the shields.

Not sure about workers though, but that could go pretty fast too. I sense an 'exploit' coming up here

[Aeson]

Try building your workers once the city reaches size 7. This will bring it down to size six, with a full food box. It will grow back to size 7 the next turn, so if you also can build the worker in one turn, the city will float at 6 with a full food box.

[zedar]

was just thinking of a further possible benefit to this.. which 'end' of the population do the workers come from? if they come from the 'unhappy' end and if 'we cannot forget the whip' unhappy people are furtherest on the end this could be a way to replace unhappy ppl because of whip with unhappy people because of population (easier to fix).
eg.
H=happy,C=content, UP=unhappy population, UW=unhappy whip.
size 6 city
H,H,C,C,UP,UW
builds worker goes to size 5
H,H,C,C,UP
jumps back up to size 6
H,H,C,C,UP,UP..

it will be good if it works like this
im still at work and cant test this though...

Zedar

[La Fayette]

IMO Aeson gives a good summary of my proposal.
I attach a savegame, so that those wishing to see how it works with their own eyes can do it.
Look at the Chinese empire. The city of Xinjang builds one worker/turn and remains size 6. The city of Tatung builds one worker/2 turns and remains size 6.
... just some kind of cheap Wonder, I tell you!... and much more powerful than many costly wonders you didn't manage to build.

[sophist]

Originally posted by zedar
was just thinking of a further possible benefit to this.. which 'end' of the population do the workers come from? if they come from the 'unhappy' end and if 'we cannot forget the whip' unhappy people are furtherest on the end this could be a way to replace unhappy ppl because of whip with unhappy people because of population (easier to fix).
eg.
H=happy,C=content, UP=unhappy population, UW=unhappy whip.
size 6 city
H,H,C,C,UP,UW
builds worker goes to size 5
H,H,C,C,UP
jumps back up to size 6
H,H,C,C,UP,UP..

it will be good if it works like this
im still at work and cant test this though...

Zedar

Happiness and unhappiness are attached to the city, not to the individual workers therein. If you pop-rush twice, you will have 40 citizen turns of unhappiness, not 20 turns of unhappiness. If you keep the city at pop 1, the unhappiness won't go away faster; in fact, it will last longer because it's spread over fewer people.

[DrSpike]

Cunning. It is certainly good to have worker factories, but I would be loathe to use any core cities for this purpose. As slow as it is, packing cities in gaps between core cities and producing a worker every 10 turns may be preferable. You can keep a steady stream of workers coming, for both pop-booming and improvements, and don't need to disband the worker factories till after sanitation is discovered.

[solo]

This is clever, but I think an aqueduct and granary are needed to get a worker every turn. The food from half of a size 7 box (left by the granary) will refill all of a size 6 box. Only 1 surplus food and 10 shields are needed for another worker the following turn. It's too bad that each one produced costs another gold to support, but I suppose any surplus workers could trot over to your capital and be traded to the AI for 30 or so gold each, as needed, to make up any deficit.
With the same sized food box up to size 6, it makes sense to wait a bit to make workers and settlers in CivIII. In contrast, CivII's changing food box made it more efficient to produce new settlers early in a city's life, since early city growth was quicker with smaller boxes.

[Aeson]

If the city is built next to a fresh water source (definitely something you should consider when placing worker factories) then there is no need for an aqueduct. A granery is needed, just forgot to mention that. Usually I forgo graneries altogether and just try to conquer the city that built the pyramids if at all possible. The time needed to build the granery in the first place (missing out on 2 units if pop rushing, 2 settlers if not) can negate its advantages later on, and if later on you can get all the graneries for free, then there really isn't any point to building them.

[La Fayette]

solo
You give a very clear explanation of the threshold effect I mentioned. Thank you. Aquaduct and granary are in fact required (except if the city has fresh water nearby,... agree with you Aeson). But I don't quite agree with the 1 gold cost: most of the time the worker will be used to grow another city and the cost will be nil.
Aeson
If one or several 'nursing-cities' are in use, no granary will be needed elsewhere, since all cities will grow with help of the workers who join them, built by the 'nursing-city'.
This is a very important secondary advantage of my proposal: cities can grow without any food surplus.

[Aeson]

The one problem I have with worker factories in general is the time it takes to set one up (granery, grow to 7), and the fact that an almost inexhaustible supply of workers is already available by military expansion (which is almost required on higher difficulty levels). I do usually set up a "mini" settler factory or two. These are set up to be able to grow to 3 in the time it takes to build the settler. Later on in the game it's worth it to try to use the 7 to 6 food box feature in settler production as well, but again it takes too long to set up in the early game, at the expense of having 2 more new cities building settlers before the first can be produced. Having just 2 of mini's allows me to continue my expansion quickly when coupled with the 1 default settler each new city with any food production will make. And later on gives me a surplus of settlers so I can raze enemy cities and rebuild instantly, avoiding any potential reversion problems. In recent games I have razed everything, excepting the AI city that builds the pyramids, and usually by the early AD's I have at least a hundred or so captured workers. Adding captured workers to cities is a little bit tricky, as going to war with their former nation can cause happiness troubles. Once I have cut down one Civ to a point were further war isn't necessary (or they are destroyed), then that nations workers all get assimilated. Usually these worker factories can produce a horseman every 2-3 turns with almost no modifications or pop-rushing needed, and not requiring a granery. Workers are great, but a single horseman unit can easily be worth 5 - 10 captured workers in many cases.

One warning, make sure to add captured workers to cities in a rotation of nationalities. Just wholesale and random addition can sometimes lead to high numbers of a single nationality in a city. In a recent game with domination disabled (civfanatics GOTM), I had a size 35 city revert to the Civ (England) I allowed to keep one city. At the time I had no idea how this city reverted, as my people were in distain of their culture, and it was 8 spaces from my capitol, and over 50 from their 1 remaining size 5 city. It turns out in my mass assimilation, this city had recieved a disproportional ratio of English workers (I think 10-12).

The point about the upkeep was intended for those situations where population points are "stored" in workers or settlers for a future time (after hospitals) when they can be re-added to the population. In this case, it certainly can be better to use settlers, as the upkeep will be half.

Also, having a surplus of settlers is always something I try to achieve, as they come in handy during wars, building temporary outposts in enemy territory for healing your troops or claiming valuable resources in allied campaigns. these towns hold the territory/resource until a more properly placed town can be founded, and then can be turned back into a settler again. It's important to remember to quickly create a culture gap between these outposts and the opposition, or rush a quick culture improvement that can later be sold. If pop-rushed, this improvement actually gives you a $ return on any excess population that may be accumilated during the lifespan of the outpost when it is sold.

Perhaps if these surplus settlers aren't going to be needed for many turns, this would be a way for them to add value instead of require upkeep. Build a dummy city on worthless terrain, and pop-rush, then sell, improvements. Even a completely corrupt city will also add 1 commerce to your coffers, instead of the settler requiring 1 upkeep. The unhapiness effects would be negated once the city is turned back into a settler as well. The one drawback would be if a garison unit was needed to keep the happiness high enough to opperate (lack of luxuries). This would also work with the production of military, but that side has already been covered. This would just be a use for those "stored" population points in times of peace.

[La Fayette]

Aeson
My experience at playing civ3 is much too short to discuss your proposals about the use of settlers.
At least I like the idea of razing enemy cities and immediately rebuilding. That one might prove very useful, especially when coming close to the core of a culturally strong civ.

[solo]

I'm liking these worker and settler ideas more and more, La Fayette and Aeson. As for early exploitation of worker/settler farms, it might be worth it to build a granary very early, in the 1st or 2nd city if it has extra food and is also located on a river. I can see this growing very quickly to the magic size of 6, where it would quickly pump out workers and settlers to speed expansion and growth.

Later in the game, nothing grows a city faster than adding workers, so these farms can pay off throughout the game. I have also discovered that sometimes adding a few workers to a proposed trade that "can't be done" sometimes tips the balance so the trade can be done, when the other ingredients of the trade are just right, too.

Aeson's ideas for keeping a spare settler or two around are excellent, too. Nothing bugs me more than seeing aluminum I need NOW just outside my borders and just inside a neighboring "can't be done" AI's, or to have a conquered city revert when it contains an assortment of my favorite attackers.

[DrSpike]

Further to my last post which generated no comment: this idea is interesting, but after consideration I feel it is too slow for early game settler churning, and not optimal for worker churning. In the early game settlers in 20 turns is the norm, or 10 turns if you take the time to build/rush a granary. This investment is usually warranted. Making the effort to carry out this plan to churn settlers fast is probably not, however, unless the map is very large. The plan has more chance of being useful for worker churning, but even here there are early game difficulties. Take emperor/deity level as an example, you aren't getting to size six without 2 MP, a temple (more investment) and 3 luxuries (I assume the granary is rushed). Probably easier is to continue with the settler churning after your expansion is complete, and build cities in gaps between other cities, and use these as worker factories. You only need two squares, the founding square and one producing 2 food - no overlap problems for ages - and you can churn workers at least 1 per 10 turns, but maybe even every 5 is the city is sufficiently close to the capital to be producing 2 shields. 4 such worker factories will keep a steady flow coming, more than enough to make improvements and pop boom later when more infrastructure is in place.

Overall La Fayette's plan above, IMO, is valuable for a midgame worker factory if the method above isn't generating enough workers for timely pop booms. The early AD infrastructural lull suggests itself.