When to Expand?

[sparkyal]

Hi, well, I've played 4 or 5 games on Prince and by the mid ages I'm getting creamed. Everyone's army is bigger then mine so everyone attacks me. This causes me not to build necessary improvements or expand. Also, most of the civs have expanded far faster then I have.

So, in looking at one of the ways to solve this problem, I need to expand the numbver of cities sooner. but when? how do you know when it is time to expand? when the computer recommends a settler (I would think not...it is the enemy)? do you experts wait until you have a certain income is a city?

So, when do you build cicites 2, 3, and 4? I'm very curious to know the answer to this question.

thanks,,

sparky

[mutax2003]

Here is my typical strategy, build a worker/fish boat in your initial city to improve the tiles, then send out settler parties until your science rate hits 60%. Each new city build a worker first, which is then used to chop obelisk, cottage spam and chop court house (but leave some plain forests around for lumber mills later on). The great library is an important wonder, so beeline for literature, and chop it out fast. After that, code of law should be your priority if you don't have it. Of course, if an opponent is close by at the start, I might forego settling a third city, instead chop out 8 axes and grab cities from the AI. Granary, barracks, courthouses, and lighthouses (only in food poor location) are necessary buildings in your non-capital cities, so build/chop/slave out, but build armies in your high production cities. If in doubt, build more soldiers, that way, your power remains strong (less likely to get attacked), and you always have options of attacking AI if the need arises.

[zeace]

The number of cities you build is highly dependent on the size of the map. But generally it's good to build your first settler or two as fast as possible. After that let your science rate guide you. If the next city you found will be marginal, don't let your science rate get below 60% or so. On the other hand, if that next city will get you some important resources and/or will secure a bottleneck it can be worth dropping your science rate to 10%. As long as you have writing, pottery, or worst case fishing you should be able to rescue your economy. There is a risk though because you need to work hard to rescue your economy so you can't be fighting a war at the same time. Fortunately, you can usually keep the AI happy for a little while.

[binTravkin]

So, when do you build cicites 2, 3, and 4? I'm very curious to know the answer to this question.

It's good to found city 2 as soon as you can, on a spot as good as you can find.
Cities 3 and 4 should come as soon as you've found sweet spots for them.

In other words, it's an early priority to scout around your starting location in search for crucial resources and then put cities to hook them up.

My usual start city placement goes as follows:
2. - a good food/production site
3. - hooking up copper
4. - hooking up horses

Any of the above resources could be unavailable in near area, in that case you'd just want to get any other special resource you can use early, like gold/silver/gems/ivory/furs/stone/marble or repeat point 2 if none are available.

It's also advised to spam as much settlers as you can if you see that you've got plenty of special resources around - that way you deny them to anyone else and make yourself stronger even while losing on research.

[Blake]

Typically I get my second city ASAP, usually after a worker, often trained at size 3, or whipped at size 4 (if I beeline bronze working). You usually want to get a worker or workboat first, but not always. Ocasionally it makes sense to go Settler first (like at size 1 as your first build), this is mainly when a worker/workboat doesn't make much sense (because you need too much tech to hook up your resource(s) and/or desperately want an early religion). Also in Warlords imp leaders can train a settler quite quickly if they can get 4 hammers right away (it involves either working a plains hill forest or founding on a plains hill).

As for cities 3 and 4, I quite often actually found them nearly simultaneously, whipping a settler from both my capital and 2nd city, there's no particular reason to do this, it's just how the timing works out, like my capital might build a wonder while my 2nd city spits out some garrison units.

Typically I'll have 4 cities by 1000BC, but I may have as many as 5, or I may only have 2 if I'm doing a hardcore slingshot or planning on hardcore conquest (usually some cities should've been captured by 1000BC, though).

You should nearly always whip settlers and workers, since once you have a granary this is much more productive than training them with food and gets them out a lot faster (Getting a settler out 5-6 turns earlier really makes a differnece in getting the best sites). Either whip them at size 4 for 2 pop, or size 6 for 3 pop, the size 6 whip works best once you have a granary, without a granary you'll get settlers out faster by whipping them at size 4. Obviously you do need high food though for a size 6 whip.

I suggest always expanding as quickly as possible regardless of research rate, as long as good city sites remain, snap them up. Only once you've run out of good city sites should you stop training settlers. You can later fill in the less optimal city sites as economy allows. The critical techs to research are Pottery, Writing and Bronze Working, if you don't have all 3 of these techs then economic recovery will be unreasonably difficult, however with them it's as simply as chopping or whipping libraries and running scientists, while waiting for your cottages to develop.

[Dis]

I try to get 4 cities up as fast as possible. I only delay 3 and 4 if barbs start popping up and I have to build more archers. This means I can't maintain 100% research, but oh well. It's necessary to get a decent land grab. I'm much more slower expanding after that. At this point I have to concentrate on defense, or the barbs will have me for lunch.

Once all the gaps are filled in and I'm sharing borders, I have a tougher question. When do I do my 2nd expansion (which of course is conquering). On one particularly successful game I wrote down the date I started my first war with another civ. 380 AD. Of course, this may change based on the world size, map type, difficulty etc. The funny thing is, I never adhere by that date. But I do use it as a guide to not wait too long. That date was pre patch though. I think I have to move things up a bit because of faster AI growth and infrastructure. Which puts their military and cultural defenses higher. But if I do it too early, I'll get hammered by maintenance costs.

[sparkyal]

Wow, this is great advice guys. I was figuring on not going below 90% on research, which keeps me with the aI (to some degree), but as you guessed it I become really small compared to everyone else and my capital is the only one that can really build the wonders, which takes time and means less military, which makes me really weak. And that is why I needed to ask what other people are doing. the advice is great! Thanks!

As for Dis, I really don't have any good advice to offer post patch...but, what I did before the patch and was good at it on the Noble level was attack my weakest and closest neighbor as soon as I had sufficient forces to insure a victory. I also wared for resources. If someone had some I wanted, I made sure to attack that city. now, post patch, military is the weakest and I'm getting creamed. I went back down to Noble and on this game I'm doing ok, but not great. wow Blake, you sure did help out the AI, great work! Now I have to learn a new way of doing things in order to win.

sparky

[Qwertqwert]

If there is a nearby AI, espically an aggressive one, I would instead of trying to build a settler quick (unless you're imperialistic) build a barracks, and a few archers or chariots or whatever youc an get and take the city quickly (hopefully before they can build a decent defense). Usaually it's good go get a worker after the barracks though, so you can produce more units faster. This also stops you from having to worry about them later. Really this is just rushing (which you'll know if you play RTSs) but I don't generally hear rushing used in Civ.

[couerdelion]

Rushing is difficult without two cities because you need some special resource to knock out a neighbours capital. The alternative is archers but these are expensive as a means to capture cities. But normally 2 cities will be enough for an early rush.

If there are good sites to settle, I would tend to expand first into this though and pick up the odd barbarian city is there are any going - it's good experience for when the real fighting begins.

But plan on taking a war to a rival AI early in the game. This will give you a greater claim on land in the area because one of your rivals has been weakened. Unless you get copper early, and rush your neighbour (2 cities then barracks and axemen), you'll either have to wait for iron (and swordsmen) or construction (for catapult). I would save 500 AD is a good guide for when you ought to have started one war.

As a rule, the timing of war will depend on many factors such as availability of strategic resources, quality of free sites, strength of neighbour, whether you are AGG or not. So there is no fixed rule.

[binTravkin]

I usually dont bother with swords at all - by the time they arrive at target, they're often late.

I try to hook up horses early and then research horseback riding - horse archers have twice the speed and are slightly better vs defending archers.

[couerdelion]

Originally posted by binTravkin
I usually dont bother with swords at all - by the time they arrive at target, they're often late.

I try to hook up horses early and then research horseback riding - horse archers have twice the speed and are slightly better vs defending archers.

They are surely not as strong as swords against archers and certainly not on a hammer for hammer basis.

Horses alone are exposed to spear attacks while swords can approach cities methodically with support troops (spears and axes). On the whole I think that Horseback Riding is a bit of a dead end tech while Swords are the primary Classical Age attack unit which come with an important military tech.

In my current game, I had my typical 6 cities (4 built and 2 barbarian) but started a war around 200BC. Actually the motives for the war are more than acquisitive because of ivory. Just like when you see the Romans have iron you may want to make a pre-emptive strike, doing the same against a neighbour with ivory before they can get Construction, is also a useful ploy.

I have yet to find out if I did attack soon enough though

[sjm]

I almost don't bother with swords either - but for another reason. CR axemen do the job almost as well, and they don't have any natural (defensive) counter. The minute the defender has an axeman sitting in his city, your swordsman is often dead meat. Not so for axes, which can take on both defending archers and axemen, with the big bonus that CR offers. Once you've got CR2 you shouldn't have too much difficulty taking cities until longbowmen arrive.

Cavalry do have the advantage that they can retreat though; a couple of flanking promotions on the odd cavalry can help soften up hill-top defenders until catapults come on the scene. You'll lose some, but you'll keep some too. But I (almost) never prioritise horseback riding - it's not until knights/cavalry that you really need it. It's one of those techs that often gets popped from a hut, traded, or researched in one turn 2000 years after everyone else has it.

[Blake]

I actually really like swords. There are two reasons for this.
Firstly they cost 40 hammers which makes them easier to double-whip, and possible to double whip if you have a forge (non-bugged whipping). Even more importantly than exploiting whipping quirks is that Swords have a MUCH higher survival rate against everything except Axemen, they live more often attacking archers, they are not vulnerable to chariots (Warlords only), they are the cheapest 6str unit, better capable of holding their own against attacking Horse Archers, Longbows and Catapults. The fact is Axemen have little longlivity outside of the early game, their main role becomes protecting elephants against spears and pikes. Any axemen heavy offense is very lossy, you can't get momentum going with axemen because you lose too many. They work fine for a quick disposable attack force to eliminate your nearest rival, but Swordsmen can actually get momentum going.

The main disadvantage of swordsmen is that they require Iron, in many cases it does make more sense to go with the shockingly powerful Eleaxepult army (if you have Ivory), or maybe a beeline to Macemen. However Swords do work well a basis for the 6str "Iron" army (swords, crossbows, pikes) - the Crossbows and Pikes provide excellent counters and the Swords provide a cheaper option for attacking cities and defending against Longbows and Catapults (I'm fairly fond of Combat Swords for this reason). It is quite odd that Longbows are actually the strongest attacking unit against CrossPike and swords are the cheapest defense against Longbows .

[couerdelion]

Even with an axeman in the city, all you need to do is through a basic CR I swordsman at the city and they will be injured. Let the stronger swordsmen then deal with the archer defenders and have one or two axemen spare to finish off the wounded axe defender and perhaps the spearman you might also find there.

Thus speaks an expert who lost 4 swordsmen the other day when attacking the Ottoman capital. One of these even got killed attacking an archer who was approaching the city. He then failed to notice an alliance that the Ottomans had agreed with the northern barbarians who sent an archer south to pillage the iron mine.

So there’s another rule. Don’t get too cocky

[frenzyfol]

Originally posted by Blake
Typically I get my second city ASAP, usually after a worker, often trained at size 3, or whipped at size 4 (if I beeline bronze working). You usually want to get a worker or workboat first, but not always. Ocasionally it makes sense to go Settler first (like at size 1 as your first build), this is mainly when a worker/workboat doesn't make much sense (because you need too much tech to hook up your resource(s) and/or desperately want an early religion).

I did a quick experiment last night
Founded on a 2/1/1 with an oasis 3/0/3 as my second worked tile. Marathon as Hannibal (fin/char).

First save I build a settler straight away and he popped out at 3220BC
So at 3220BC I had a size 1 city that had not started to grow yet, and a single settler, and the 1 warrior I started with.

Second save I waited until I had just popped into size 3 before choosing settler. Previous to that I had managed to build one and a half warriors. I was working an oasis 3/0/3 and a flood plains 3/0/1 and bee-lined for bronze (slave).
I pop rushed the settler at 3025BC
So I was 13 turns behind my previous save, but! I had a size 2 city and two warriors, also I was slightly ahead in beakers (didn't measure how many, might do that tonight)
I'm pretty sure you'd need to start with mining to pull this one off.

In conclusion I'm still not entirely sure which is the best strategy. Having that size 2 capitol would allow you to pop out another settler even quicker, allowing you to catch up with the settler first strategy.

Has anyone done similar experiments? What did they find?

[Arrian]

I tend to go worker-first and then grow to the happy cap whilst getting bronze working (obviously not beelining) and then whipping settlers out at happy +1. I do not invest in a granary right away. I'm sure Blake can demonstrate that it's suboptimal play in more ways that one.

I find that I want my city to take ~10 turns to grow back the pop I rushed off (say happy +1 is size 7. I whip for 3 pop, dropping to size 4. In a perfect world, I want it to hit size 6 10 turns later, grow to size 7 fast, and whip again) and in the meantime I'll be building other stuff - military, library, etc. Of course, if I build the granary and do some micro managing (including specialists), I could do that better than going no granary. But I'm also trying to fit in an Oracle CS sling, so my build priorities are all over the place.
...

My most effective rush of late was a mass catapult rush. I had a couple of axes and a spear for cover (though I did replace the spear with a WE once I got ivory online). True, Construction is not really "early." But in this particular game I had very strong early research (5 cottaged floodplains + 2 gold mines between my first 2 cities). I built only 4 cities, scraped together about 8 cats, 2 axes and 1 WE and unleashed hell. Even if you replace the WE with a spear (and I had built one), my AI target could not have stopped me. I took very light casualties and absolutely crushed him. Catapults are really powerful. If one lacks a powerful early UU (WC, Immort, Praet, etc), I think they're the way to go. Hurry to construction and punch out ~10 of them, cover with a few counter units and go bonk some heads. Adjust for level as necessary, as YMMY

-Arrian

[zeace]

If the next tile you will work will add a significant number of food/hammers, then it's probably better to grow before you build a settler/worker. If your next tile is going to be a forest or something, you're better off building it now. That means that if I have a great capital location, I'll hold off on building the settler until my capital is big. But if my capital is marginal, I might go settler first.

[Olavi]

Four cities is usually my early expansion target. I try to get those four up as soon as ever possible. I usually play commercial civs and prioritize cottages, so economical reasons rarely force me to slow dowm my expansion.

After that, expansion depends on many things. If there are still sweet spots around, I may go for cities 5 and 6 quickly, too. If not, I may start developing my strike force and take cities 5 and 6 from a neighbor a bit later.

I've got a question, too: is there any means to know the maintenance cost of another city before actually founding that city?

[couerdelion]

It seems to me to be very situational but I’m loathe to offer much in the way of advice because my more recent efforts seem to demonstrate that I have forgotten how to play this game. Certainly the higher levels will leave you far less room to expand so there are two options

1) Hurry to grab the good sites
2) Get the strategic resources and then steal your rivals’ cities.

The second option is likely to be the cheaper means of acquiring a city - or at least relatively cheaper because your rivals suffer a loss. As long as you can get one of the strategic resources, option 2) will always be open to you.

But if you’re rivals are some distance away, then this will not only require a longer time to build up and move your armies but will also leave you will a rather disjointed empire that requires more units to defend. In this case, it may make sense to continue expanding until you meet and then consider expansion in another direction or move onto the offensive. In this case, unless your rival has some very juicy sites, I would opt for continued expansion into areas with some good resources or try to capture barbarian cities first and figure out an expected limit on natural growth so that you can plan for your first conflict.

Of course if you have Romans nearby and they have iron then you may want to go to war early. Also if your neighbours have ivory then you might also want to strike before they get construction. These are examples when you need a pre-emptive war which, in my book, over-rides any personal desire for peaceful expansion.