Settler First Build

[drkodos]

I feel this tactic is mostly overlooked and I have oft wondered why. Building a Settler first seems a controversial starting tactic and has not been accepted by too many players as a viable starting strategy, but there has never been offered a clear refutation.

I believe and argue here that it remains a strong starting tactic, is highly underrated, and ignorantly ignored. Nothing can accelrate an early land grab as fast as getting that second city built 14 turns into the game. And certainly, with Imperialist trait, it is even more powerful and fast. The primary advantage of such a seemingly counter intuitive start is that it improves the entire empire immediatly by effectively doubling production early, and thus allows for faster early game growth. It also is tremendously flexible because you can mix and match production from your two cities while almost every other civ would still have only one city and can only pump out one item. With settelr first, you can have a worker coming out while the other city builds warriors all while your opponents are still growing their first city.

It's all about the real estate. They ain't making any more of it, and the sooner we make claim to it, the stronger we are.

It is rare indeed that the AI will launch an offensive prior to 2,000 BC, but it is not uncommon for that same AI to have snagged a couple of good city sites close to our home territory by that time.

Because the game is essentially about working as many tiles a possible, nothing facilitates this as quickly as getting that 2nd city site up and running and nothing accomplishes this faster than building settler first.

It is a very strong opening salvo in games in which starting positions are not ideal and there is some better land to be had proximal to a starting position. In games in which religious techs are being pursued and there will not be too many early worker chores, it is particulary strong.




Comments, criticisms, caveats & condescension welcome.

 

[wioneo]

I usually hold off on founding cities until there is something worth founding a city for. Iron, copper, or horse, for instance. I can't see myself waiting until iron working, but animal husbandry and bronze working are soon enough.

 

[abuaftab]

This is my normal strategy. I never wait except for a worker if I have picked a civ that has wheel first off the bat.

 

[Wodan]

I usually hold off on founding cities until there is something worth founding a city for. Iron, copper, or horse, for instance. I can't see myself waiting until iron working, but animal husbandry and bronze working are soon enough.

Seems to me this is irrelevant to the OP. The timing would work out for your 3rd city, instead of your 2nd....

Wodan

 

[DaveMcW]

The typical city gives 2 2. A worker can give you that much by improving resources almost as fast as a settler, plus the worker keeps on building more improvements faster than a city can grow.

Settler first is a valid gambit if you know land will run out fast, i.e. deity level or against a human choke-rusher.

 

[OceansEleven]

No... I hate it.

I'd rather be building a warrior *something you NEED* while at the same time growing the population, hopefully to 2 by the time you finish. Then you build something else you need, maybe the settler.

I usually build a worker first. Nothing beats getting bigger quicker, and thus slowing down your settler build.

 

[VoiceOfUnreason]

Nothing can accelrate an early land grab as fast as getting that second city built 14 turns into the game. And certainly, with Imperialist trait, it is even more powerful and fast. The primary advantage of such a seemingly counter intuitive start is that it improves the entire empire immediatly by effectively doubling production early, and thus allows for faster early game growth.

Thoughts....

One, maybe I'm doing it wrong, but it normally takes me longer than that to get a settler trained.

Second - double production isn't something I normally care about at that stage of the game. Assuming ancient era starts, and so forth, I don't generally have two of anything that I want to build yet.

Plus, as it appears to me, worker -> farmed food resource is a lot closer to doubling production in a real and practical way, and happens 5 - 10 turn sooner.

 

[miller4242]

Settler First???? Intriging....

Here's one point to consider. Build settler first. Settler founds second city. Then look at your GPT. I believe in most cases it will take a hit, which could possibly have an impact on your early research, at least for a couple of rounds.

I agree with the point about wanting to build a worker first to improve your capital. However, here's scenario that needs to be considered if you build a settler first. What if, at your first encounter with barbs, your scout gets whacked. You would not have a good layout of your nearby lands, and would not have sufficient info on how to plan your empire; ie. where to put your 2nd, 3rd, 4th city and so on. Plus you will not know where your opponents are at.

 

[drkodos]

Thoughts....


Plus, as it appears to me, worker -> farmed food resource is a lot closer to doubling production in a real and practical way, and happens 5 - 10 turn sooner.

No doubt an early farmed resource is a worthy tactic, but there are many start scenarios in which food resources are not within the initial city site cross,and there are many starts in which a worker would not have suffiecent jobs to do to warrant their immediate construction (i.e.pursuit of religious path)

Be mindful I am not advocating this as an all-the-time tactic, but one that is highly effective across several common starting scenarios. I would say on average, in about 20 to 25 percent of all starts, it is the strongest possible tactic to use, especially on levels Monarch and above where real estate is pricey and there is chance of getting crowded out quick.


In games in which the AI starts with a settler of its own, it is most useful to get that 2nd city up as quick as possible. The more land we own, the higher the liklihood that a strategic resource will pop up in one of our tiles. Then, for the 3rd city one can either follow the resource trail, or bum-rush the opponent to assimilate their lands into ours.

Most, not all, games are won by the player (or AI) that controls the most tiles. Of course, OCC and other esoteric cases need not apply here, so let's not get bogged down in arguing over those instances because it should be apparent I am not referencing these types of game conditions. Tile control is the key to the game. Controlling more tiles is the best thing one can do to ensure winning chances and nothing controls tiles better than a city and its cultural borders. More tiles equals more resources equals more food equals more of everything we need to win.

So, with that thought in mind, settler first jumpstarts every winning condition.


As far as doubling production, two cities can pop two units simultaneously and that is something one city can never do. And that is the early crux.

 

[VoiceOfUnreason]

No doubt an early farmed resource is a worthy tactic, but there are many start scenarios in which food resources are not within the initial city site cross

That's rare to my experience, which is admittedly finite.

Be mindful I am not advocating this as an all-the-time tactic, but one that is highly effective across several common starting scenarios.

It might be useful to discuss one or two of the most common of these scenarios in detail (perhaps with examples). Sounds like a religious gambit would be one of those. Can you roll up a start that's appropriate for some scrimmaging?

 

[drkodos]

That's rare to my experience, which is admittedly finite.



It might be useful to discuss one or two of the most common of these scenarios in detail (perhaps with examples). Sounds like a religious gambit would be one of those. Can you roll up a start that's appropriate for some scrimmaging?

Well, I just shadowed Sisiutil's ALC game where I thought it was a doable variation and I was intending to post up results as soon as he gets his game done as I do not wish to spoil it. In fact, it was that very ALC game that compelled me to post this thread, so I will hint that it was a highly successful endeavour in that instance.

 

[Gnarfflinger]

It might be an interesting gambit to try with a good rushing civ like Rome or Egypt. Three cities when you would have 2 means more of the rushing unit...

 

[VoiceOfUnreason]

Well, I just shadowed Sisiutil's ALC game where I thought it was a doable variation and I was intending to post up results as soon as he gets his game done as I do not wish to spoil it.

OK, that's a hint; but he's likely to be a while, so perhaps rolling another start in the mean time?

 

[drkodos]

OK, that's a hint; but he's likely to be a while, so perhaps rolling another start in the mean time?

Fair enough. I was planning on getting something up early this week (Mon or Tues). I was going to do the ole side-by-side comparison deal using the same start and playing it out for 50 turns with several different starting tactics so some type of legitimate comparison could be made. So, if I do not get banned again, I will be glad to accomodate and hopefully get a bunch of input from people like yourself whose opinions are most respected.

But for now, it is goodnight as it is 1 am EST and my eyes are getting ready to go on strike.

 

[mice]

Sounds good. Roll a start and people can play with a build order of their choosing for 60 turns or so. Then a comparison of prod,tech commerce can be made, amoung other factors.

I'd like to play warrior(or scout)/ worker / settler,
because I always play that ( but have the feeling it's sub-optimal in many cases)

 

[Charles 22]

I think I will at least try this idea. My game generally evolves around starting a brracks and then when I get pottery switching that to a worker, as cottatges are my first priority, though I don't beeline research that way so much. After the worker tha barracks completes and then I get 2-3 archers and then the settler (I don't build any help for a scout I may have until much, much later, if then). Doing as suggested would speed my settler probably at least 150 turns. The key here is that I could ignore building the worker for example on city one real early, and have him roading or something between the cities, or even have the 2nd city be the first with the barracks.

I'm not sure it will change my overall strategy, but since my 2nd city was geared more for the best river location anyway, what is revealed later matters little at that point anyway. I think I will end up delaying my 3rd city, and having it wait on more resources revealed, but with 2 cities up earlier, it ought to make up for the 3rd city delay, and with 2 cities up earlier I may accumulate enough research more quickly to actually have the 3rd city coming out earlier than it did as well, even with waiting on more resources.

I will definitely try it anyway. I do somewhat slobber at the notion that I will have a city up so early that doesn't have to worry about making barracks, although I will caution that with the Germans I'm usually playing, that first settler is at least a 30-40 turn wait.

 

[tazpn314]

As with all strategies it depends on difficulty, speed, map size and resources. I normally play marathon or epic speed and it normally takes 25 turns to spit out a settler.

If I have food plains available (or workable food resource) in the 9 square radius I will normally work the plains first which takes 8 turns to go to 2 citizens. At that point the settler cost is now 20 turns and now I have an extra boost to research and a little extra gold as well as being 8 turns into building a unit or barracks for the cost of 3 extra turns before the settler is ready. The pump is then primed if I want to immediately go for a worker. Also that extra citizen is available for slavery if I'm lucky enough to get BW quickly.

 

[aelf]

How about the potential loss of a goodie hut or two that you may find with a 2nd warrior/scout? That is pretty random and unreliable, but it should also be considered.

Personally, I like to build a warrior/scout first to help explore the map. It's also in case the 1st warrior/scout dies to animals/barbs. Either that or build a worker first.

 

[WilliamOfOrange]

My game generally evolves around starting a brracks and then when I get pottery switching that to a worker, as cottatges are my first priority, though I don't beeline research that way so much.

So you don't build a worker until after you have pottery? What about timing it to be build the turn you get pottery? Doesn't that seem more economical?

What I often do is have a building in the queue and then once I hit size 2 or have my borders expand and gain a better food source, then I switch to worker or settler. Doesn't that expediate the settler or worker production?

 

[asabahi]

I can see the settler first approach if you are playing, for example as Monty, where your starting techs are Mysticism and Hunting. Since it is likely I would try to get an early religon with him, why not build a settler first and get a second city built early.