I don't think it makes sense make a coastal city on marginal land just to make one that uses coast - I would much rather put a city in the jungle than either of the blue dots in the other plans - the jungle is a pain, but at least with our army of workers, it can be improved.
Well, the problems for settling an inland city, are:
Have the wheat as food, horse, 6 hills (3 grass, 3 plains) and 6 peaks, if you plant it for all 4 grasslands. If you move the city site then it loses grasslands; the 'best' production site for an inland city prior to windmills is only capable of 25 hammers per turn without a forge (a few more working irrigated plains), and isn't capable of making any commerce, whereas the caostal site that I marked can make 24 hammers per turn but is also capable of making commerce (this is going to be very important while we are expanding; the less commerce we generate the less cities we can build). It also doesn't leave any space for a coastal city.
If you plant a city on that coast without any food, it is going to take forever and a day to become productive. We also need to have a bunch of tiles that generate 2 food per turn to continue growing, coast does this nicely, and pays the bills, so it'as actually quite good. And if we build the Collosus, then the coastal cities can pay all of their maintenence and the maintenence of a bunch of other cities as the coastal tiles each make 4 commerce. A bad coastal city without food though can't do this because it has to work the irrigated grasslands to grow and so is a drain on the economoy is it isn't making any commerce. An inland city is in the same situation: while it may be able to grow, it can't make any commerce, and so we have to limit our expansion.
So, the comparrison isn't just between the city sites, it's also between what other city sites we could settle before we got near to crashing our economy. Two good coastal cities plus the Collosus would allow us to settle up to another 2-4 cities if we got the coastal cities down first and started them growing, whereas one coastal city and one inland city/slow growing coastal city would only allow us to fund at most 1 other city before that group of cities was a net drain on our commerce. Coast, in this instance, is more useful that a bunch of plains hill that we don't have the food to support.
The current CDP plan that I posted doesn't have us getting any more workers until after t76, as 6 workers are enough right now to keep everything on track, but when we start expanding properly, we are going to need to most likely at least quadruple the number of workers we have in very short order, and until we have done that we don't have an awful lot of slack in the worker movements, and jungle is going to eat up a large amount of our workers' time. We have jungle at JD that needs to get chaopped for health reasons and to make more cottages, which shoul IMO be a higher priority that another jungle city (which would be second in line IMO, as JD needs to be working cottages as much as possible in preparation for Bureaucracy). One of the advantages of getting the collosus at sheep is that that city is capable of 3 turn workers in a hammer configuration, so between sheep, CK and IS we are capable of making all the workers (3 turns in sheep), settlers (4 turns in CK) and military (2 turns axe/spear in IS), which would help our expansion an awful lot, as otherwise we need to fit workers into IS and CK which slows our expansion.
---
The Arguments against settling 2 cities over there early though are these:
- No horses. Counter argument is that we have horses down south, just plant a city for them, as we'll want a city down there anyway, and it's gotta be better land than we have to our east.
- Maintence, we would be paying quite alot for cities over there. True enough, but with the Collosus (plug, plug, plugging away) they would be a net increase in our GNP, which couldn;t be realised by an inland city and would be realised alot slower by a coastal city without food.
Corroloary of this: we would be delaying settling a cottage site such as the jungle if our maintence was too high, and we would have to get out another settler. Yeah, we would need to get anotehr settler, but we would also need alot more workers to improve a jungle cottage site than we would to improve a coastal city with one source of food; we would be sending two workers over to hook up the gold, we wouldn't really need more than that to keep both improved as both cities would be working the coast to grow anyway, and that gives us enough time to improve the land with just 2 workers.
- Can't take Moai into account until we have scouted all of our coast line. True, but we need to get boats out to do that well, and there aren't many place that could make use of Moai as well as clams; for instance sheep only has 7 coastal tiles, so would be better suited to being the GP pump.