Important GA cities...

[Aeson]

We are going to have 3, and possibly 5, Aquaduct cities by ~700BC. Here is an estimate of what we can expect from each of them as far as production and improvements go. The numbers are assuming we trigger our GA. Looking at either of the scenarios in the GS Scenario thread can help visualize the impact that each of these cities makes on their neighbors. In some cases it's almost nil (EotS as described, Hurricane), and in others it really makes a difference.

If we should miss the Pyramids... there are always WC's. We could better time our GA then, but miss out on all those free Granaries.

EotS - At size 9 we can be producing 20 shields per turn here. Temporarily we can boost this to 25 as would be helpful in building a Marketplace, but food would be an issue. Good uses for a 20 shield per turn city are Barracks + Spears/WC's, Improvements, or even a Wonder. The alternative is to continue to flood with 2 turn workers, though we can actually keep growing slowly at the same time.

Hurricane - At size 12 (Irrigated Grassland, Mined Mountain and Iron) we hit 30 shields per turn. This is the best site for a second wonder, but with a Barracks already it is also the best site for Horsemen/Swords, and with a Harbor build it's also the best site for Galley production. Not to mention the possibility of building Improvements here.

Tempest - At size 12 we hit 30 shields per turn here as well, but it may need 3-4 turns into our GA to reach this point. Possibilities for this city are much the same as Hurricanes 'land' ones.

Possible Aquaduct cities:

Spinebreaker Ridge city N of Tempest - Built on the river, this city would come with an Aquaduct free. It will have a Barracks before the GA. At size 10 with a Courthouse (and 3 mined Mountains.. ~ halfway through our GA) it can be producing 20 shields per turn, but it is a much better cannidate for 15 (Size 9, no Courthouse, 3 Mined Mountains) or 10 per turn. Almost certainly should be producing units, but some Improvements (Library, Marketplace) inbetween would be useful if we aren't specifically gearing up for a war. Also a decent cannidate for a Middle Ages Wonder prebuild, or even our FP (for later Leader rushing our Palace elsewhere). In the case of a wonder prebuild, a Library or Marketplace could replace the Barracks build and we could take our time building it up (those mountains take a lot of worker turns).

Cyclone - One of the coastal cities would be better for an Aquaduct in terms of growth, commerce, and naval capacity, but if we need another production powerhouse (20 shields per turn) this is probably the only other spot for it for quite some time. Any of the coastal cities just won't have enough shields without cannibalizing their neighbors. The best site for building Workers (after our Settlers are all produced) to help build up the populations and terrain improvements of any of the above cities, in which case an Aquaduct isn't necessary.

Useful non-Aquaduct cities:

Tornado - If we want a 15 shield per turn Galley producer, this is the site. Doesn't even require an Aquaduct, just to not chop the Forest N of it (better than giving it another Shielded Grassland from another city, seeing as it will be sitting at size 6). Could be ready by the time our GA hits, but takes a Barracks out of the unit rotation.

SW of Cyclone - Hit harder by corruption than Tornado. 10 shields per turn (size 6) would be the target here. Even that makes the city E almost shieldless though.

-----------------

Any other throughts as to major production sites we should try to set up by around 710BC?

[nbarclay]

What we may want to do with Cyclone once we build the Pyramids is transfer its high-food tiles to other cities (which can replace it as worker producers) and focus it on building an aqueduct. Under Republic, we'll get more food from our high-food tiles than we do now, so our options for where to build workers will be wider.

Anything farther away from EotS than the cities we have now probably needs to make a courthouse a high priority, although the one we're about to build (Typhoon?) needs a temple first so it can work the whale tile and the southern fur forest.

[Arrian]

Aeson,

Are you considering corruption (or waste, rather) in your calculations of shield output?

-Arrian

[Aeson]

Yes Arrian. I'm basing it off actual results of testing in the scenarios. I recommend everyone downloads them and plays around, it's a lot of fun! Not to mention it gives a very good feel for where we can be at that point.

-----------

I played around with swapping the Wheat tile from Cyclone to the coastal cities to facilitate faster growth, as even producing 2 turn Workers Cyclone will fill up size 6 too fast using the game and wheat.

Cyclone probably is the second best place after Tempest for an Aquaduct. If we take 2 or 3 turns off research (either all at once or peicemeal), we can basically cash rush the Aquaduct there, and the extra Workers we can get in the interim will pay off the expense rather quickly. If we just build the Aquaduct normally, even taking away all Cyclone's food sources more than 2 will still have it sitting on size 6 for much too long.

I'll look at the other cities and see if we can't set up another Worker factory or two at least temperarily. Until we max out the population of our 'good' cities, the 10 shield expense of transfering pop will almost always pay off in less than 10 turns. Especially considering that growth would be stopped in the original city otherwise.

Courthouses are a possibility. The Spinebreaker city and SW of Cyclone being the two best targets. Aquaducts in general will do more for overall commerce and production given our city layout though.

A Courthouse might make the same pop more productive (varies a lot), but the Aquaduct allows the pop to grow to double. It's closer than that sentence makes it sound, as a Courthouse gets built earlier, and doesn't require the time for pop to grow.

[nbarclay]

Note that a lot of our cities are going to need to work water tiles instead of land tiles as they go past size 6 (especially if we get EotS and Cyclone hogging land tiles), so aqueducts won't help their production, just their gold. Further, they'll need harbors to make much use of the water tiles they need. I think that argues in favor of courthouses, which in turn speed up production of aqueducts and harbors. Eventually, we'll want all three, and building the courthouses first will be the most efficient order in terms of shields.

Nathan

[Aeson]

The inverse is that waiting for a Courthouse then Aquaduct limits our growth, and so once both are built (in whatever order) the Aquaduct first city will be producing more at that point until the Courthouse first city can reach size 12. Even then, if we pop out a Worker or Settler (and have something to do with it) every time the city reaches the size 12 limit, we can maintain the advantage. This pop advantage can last the entire game if Conquests go well, but as translates into production and commerce, will only last until all our productive cities are maxed out in pop.

At some point corruption does make an Aquaduct first too time consuming, and the population advantage nullified because the population will be corrupt. I don't think any of our cities are there yet. The northernmost city looks to have about 50% corruption, and a some of that is distance rather than OCN. Normally I'd say the OCN is a good break even point, but as you say, the terrain available plays it's part as well. Definitely something that should be looked into.

As for Harbors, at least in our most productive coastal cities (the ones most likely for Aquaducts and Courthouses) are probably better off Harbor first so that we can have Vet Galleys.

[nbarclay]

A lot depends on how soon we expect to get involved in heavy fighting. If we expect to do a lot of fighting soon, long-term investments are of limited value. If we plan to wait a while to do much fighting, long-term investment is more important.

[DeepO]

Indeed, planning our fighting is important, but isn't it so that the best, soonest moment to do our fighting is at or near the end of our GA? Maybe we can wait a while to get to knights first, maybe we don't want that, but in any case the possible WC-boom in a GA makes a lot of things possible.

courthouse - aquaduct order: it will depend on where the city is built, but in general I agree an aquaduct should be first if possible (i.e. not too much waste). The faster those cities grow, the better... we'll need workerpumps to keep our tiles improved, though. Is there any chance of making a 1-turn workerpump? Cyclone could do it, but perhaps at too great expense...

DeepO

[Aeson]

I just looked at Cyclone again, and I really like DeepO's suggestion. We can set up Cyclone at size 4 or 5 to churn out one turn workers throughout our GA. The only problem is it might be a bit of population overkill... but if we hit Vox all those extra Workers will pay off quickly.

[nbarclay]

What I had thought about doing was shifting worker production elsewhere so Cyclone can build an aqueduct, since it's one of our few inland cities and therefore capable of higher production than most. Bolderberg, for example, could alternate two-turn WCs and one-turn workers, although it might have to work a hill to get the needed after-corruption production. We could probably do the same sort of thing in Tornado. That would sacrifice some militry production in the short term, but having Cyclone stay at size six costs us quite a bit of long-term production.

[Aeson]

The 1 turn Workers would help us build up EotS to 30 sheilds per turn as fast as possible. Maybe we could start the Aquaduct after that point?

The impact of building up Cyclone on Bolderberg, Typhoon, and/or Tornado will be pretty big though. I'd prefer to keep the 1 turn Workers throughout the GA, and then pump up Cyclone afterwards. That way we can keep two 15 shield per turn barrack cities going, and keep Typhoon building Galleys at 10 shields per turn. Maybe use the end of our GA to get Harbors in the surrounding cities at least, so they won't start starving if we take all their land tiles away.

Some of the extra Workers Cyclone produces in this manner could then be added back to Cyclone once the Aquaduct is completed. That will basically double our food production in the city (10 food workers being added as 20 food pop points) at the cost of 10 shields per pop. The new citizens in Cyclone could pay back their 10 shield investment rather quickly.

[DeepO]

I like it most because 1-turn workers could be used in cities with fewer growth. Adding them to Cyclone after the GA would certainly be good as it creates a industrial city, but there are quite a few harbor cities which will only grow each 10 turns after an aquaduct and with the pyramids. If you add workers to those cities, you basically get 2-3 gpt per each 10 shields invested... If we want to swap shields for commerce, this should be one of the best ways of doing it, even better then the Collosus cost wise.

DeepO

[DeepO]

Come to think of it, as long as we don't hit maximum population in our empire (this will happen rather quickly), the 1-turn worker plan is always better then building the Collossus. The best we could get on the wonder would be 7 turns (in which Hurricane would lose food, instead of growing), in those same 7 turns, 7 workers could create 21 commerce before corruption, which would be more then the maximum 13 from the Collossus, at only 70 shield cost, instead of 200.

Nobody prevents us from going for both, though

DeepO

[Arrian]

Mmmm... Colossus. Big Bronze Statue of Man!

But the worker pump sounds good too. We could use swarms to mine all those hills we've got and then boost the cities up there to max size w/o waiting for them to grow on their own.

-Arrian