Opening Gambit

[barefootbadass]

I had this idea last night , and haven't tested it out, but you could speed up the time that you produce your first settler at the beginning of the game by adding your worker to your initial city. This would reduce the time from size 1 to size 3 in half, or from 20 turns to 10 turns in an average position.

Obviously this would be less advantageous with an industrious civ, although it still might even be worthwhile then. Also, you would also be probably working on the settler from the beginning, or only making at most one warrior before the settler. This would be less risky to do then, with an expansionist civ, which can do some exploration for where the settler should go and assure that the settler doesn't run into a barbarian or something. An industrious civ could build one quick square of road to explore a bit and then add the worker back, which would gain one turn back speeding the settler, or otherwise just keep the worker.

Basically, depending on how close the cities are, you would have the extra city about 6-8 turns earlier than you could otherwise, depending on how far you walked the settler. You could then catch up on cranking warriors, and use one city to make another settler and the other to 'pop' the worker back out.

I don't know if anyone has thought of this yet, although I did have a sense of 'deja vu' when it occured to me.

[Lord CocoaBeans]

I have already thought about this idea and was in the process of play testing it before posting it on the forum but since you bought it up I though I better add my 2 cents, this tactic works extremely well some of the time but other times it can make your life a living hell.

In order to use this tactic effectively you must balance the long term benefits of early improvements with the short term bonus of a population boost. More info once I play-test this some more

[barefootbadass]

Originally posted by Lord CocoaBeans
this tactic works extremely well some of the time but other times it can make your life a living hell.

Yeah, I thought it would be that way, that's why I called it a gambit. Anyway, I'll test it as soon as possible, may be a while since school is starting back soon and I play slowly anyway(I've yet to see the modern age or even finish a game). Anyway I'll be looking forward to any reports of people who beat me to testing it

[Aeson]

A few problems that I see in this opening, some of which you also mentioned. One, you wont have a road to your next city site, losing 2 or more turns from the advantage in getting your second city founded (the whole point). Also it cuts out the 2 warriors that would be out exploring earlier meaning you will need to find the best city site with your settler, which may cost several more turns. Another is if any 3 food squares are available, it cuts into the advantage even more. This won't work on Deity either, as you need a garrison unit or a luxury hooked up to avoid disorder at size 2. Im not sure about emporer, but it might also have the same problem. Also, if barbarians are enabled, it is much more likely to lose your settler before the second city is founded, or at least have to found it in a place that is less than optimal. Finally, there wont be that free worker mining grasslands and building roads for trade still going...

The only time I could see this helping is on difficulties where building a settler without garrison units is possible, and when pop-rushing conquest is the main goal. Also using expansionist Civs would increase its worth. Even then it probably would be better to add the worker to rush a military unit instead of building a settler if warrior code (or the wheel) was available. An archer, bowman, or even chariot (if youre really lucky with the japanese) at 3850BC has a very good chance of finding and capturing an AI city on most maps.

Come to think of it, a city site with game in a forest (2 food 2 shields) might be a situation that would make this opening worthwhile. With 2 game, that would have your city at size 2 with 10 turns till it reached size 3, and 5 shield production. That would allow for 2 warriors to be built in 2 turns each (4 turns), and a settler in 6 more turns (when the population would reach 3 anyways). On Deity the first warrior would be built in 4 turns though, as adding the worker would cause unhappiness, and so only 3 shields would be produced per turn either way. Adding the worker after the warrior was built would put the settler at 6 more turns, with the city growing at the same rate. This would still leave your settler with the problems of finding a city site, and not running into barbarians if using a non-expansionist civ though.

[absimiliard]

I believe I can offer a suggestion to optimize this gambit.

Your main benefit is not in increased production from an extra population early on, the main benefit is in the quicker jump from size 2 to size 3 and thus a quicker settler.

Given this you should use your worker to work on the surrounding area until your city pops up to size 2. Only then do you add the worker back in, and then as soon as possible build your settler.

So I'd venture the correct timing is 10 turns of worker building improvements while your city builds 20 shields of settler. Then build the worker back in and you should finish the settler fairly quickly, thus gaining turn-advantage.

This way you get some initial improving before you give up the worker, ten turns will normally get you a road & a mine. If you're industrious you'll do even better with it. Your turn advantage looks to be a gain of roughly 6-8 turns.

The loss as I see it is time it takes you to build a worker to replace the one you just spent. If someone can balance that against the turn-advantage we might be able to quantify the question of whether or not the gambit is worth it.

[barefootbadass]

Originally posted by absimiliard
The loss as I see it is time it takes you to build a worker to replace the one you just spent. If someone can balance that against the turn-advantage we might be able to quantify the question of whether or not the gambit is worth it.

Well basically the immediate advantage, the 6-8 turns of having 2 cities when otherwise you would have 1 is mainly that your food gathering rate is doubled. Production is roughly doubled too, but the food thing I think is more significant, since that is what will replace the worker and allow continued expansion. These doublings are temporary, but they also make you closer to your next settlers and or workers. The food advantage declines over time, but the jump start in your core low corruption cities may be worthwhile.

[Lord CocoaBeans]

I've just had a new idea on how to use this strat effectivly. For the purpose of this strat assume i play the Americans (expansionistic, industrious) and am surrounded by mainly grassland. As soon as I've plonked down my first city I begin to build a warrior while my worker starts to build a mine on a grassland (preferably one with a shield) once the warrior is complete I build another one (meanwhile my scout is exploring as much as possible) and my worker mines another square. the second warrior, the mine and hitting population two should happen roughly at the same time. I now fortify the second warrior, move my worker to the city and start on a granary. Next turn I add the worker to the population and rush build the granary. this does a number of things; 1. I can now continuously build settlers one after the other. 2. I have two warriors on garrison duty so the town isn't going to collapse into unrest becuase of forced labour 3. have significantly sped up my early expansion. after that my second city would be built apon a shield rich area (e.g. forest) and would start to pump out warriors/spearmen to garrison the towns that are now sprouting up like wildfire.

more info on this idea after I play test it a bit more :-)

[Dimension]

CocoaBeans:

You suggest making 2 warriors and a granary before you found a 2nd city or make another worker. While that isn't TERRIBLY slow, it seems pretty ridiculous to get rid of your only worker, too. I'm all for using population to rush granaries and temples, but not at the expense of settlers and workers. The worker will obviously need replaced, and the population you lose when you build a new replacement worker will offset your city growth just as much as the growth you thought you were gaining by building the granary a early in the first place -- and that's without even considering the infrastructure you would've build with the worker in the time it took to replace him.

[Dimension]

I must admit that I've only actually played two games of Civ3 to completion, but I have played the opening few turns many many times to get a feel for it and develop specific openings that I am comfortable with.

The first few turns of a Civ game are actually a little stressful for me, because I am constantly aware of the fact that the smallest thing in the early game will save you a few turns which can make a huge difference later.

Anyway, I have never been able to find a situation where it was beneficial to add my first worker back to my city. I've tried all kinds of scenarios, even going as far as lumberjacking a forest and adding the worker back to produce the first settler in 9 turns.

A good way to compare various openings is to see how long it takes you to have 2 cities, 2 warriors, and 2 workers. I'll assume you're playing Monarch or lower. The same principles apply on Emperor/Diety, but since you need a garrison to avoid disorder at size 2, you have even more time for your city to grow, making it that much more obvious that adding your worker back to the city is a bad idea.

For my example, I'll use the Egyptians, getting a good start next to irrigatable wheat and a grassland (shield). First, try it with adding the worker back to the city to make the settler faster:

4000 BC (00) founded Thebes (work grassland), start settler, move worker
3950 BC (01) start mine on grassland (FFS)
3900 BC (02)
3850 BC (03)
3800 BC (04) mine finished (FFSS), start road
3750 BC (05)
3700 BC (06) road finished (FFSSC), move to city site
3650 BC (07) start road
3600 BC (08)
3550 BC (09) road finished, add worker back to Thebes
3500 BC (10) (Thebes working mine+wheat, one entertainer)
3450 BC (11) settler finished, Memphis founded 2 squares away

3400 BC (11) Memphis=mine, Thebes=wheat
3250 BC (15) Memphis produces WARRIOR
3200 BC (16) Thebes switched to working non-mined grassland for 3 turns
3050 BC (19) Thebes produces WORKER, Memphis produces WARRIOR (Thebes=wheat)
2900 BC (22) Memphis produces WORKER

That's pretty quick... 2nd city founded in 11 turns. It takes 22 turns before you have 2 cities / 2 warriors / 2 workers, and you have some roads and a mine.

Now, we do it without adding the worker back to the city:

4000 BC (00) founded Thebes (work wheat), start settler, move worker
3950 BC (01) start irrigating wheat
3900 BC (02)
3850 BC (03) irrigation finished (FFFF), move to grassland (FFS)
3800 BC (04) start mine
3750 BC (05)
3700 BC (06) Thebes=size 2, working grassland+wheat
3650 BC (07) mine finished (FFSS), move to space between city sites
3600 BC (08) start road
3550 BC (09) Thebes stops working wheat, working mine+grassland
3500 BC (10) Thebes=radius 2 (reassign mine+grassland), road finished
3450 BC (11) start road on future city site
3400 BC (12)
3350 BC (13) settler finished (start WORKER), road finished, Memphis founded

3350 BC (13) start 2nd mine on grassland, Thebes=wheat, Memphis=1st mine
3200 BC (16) mine finished
3150 BC (17) Memphis produces WARRIOR, starts 2nd WARRIOR
3100 BC (18) Thebes=size 2, Thebes=wheat+mine
3050 BC (19) Thebes produces WORKER
2950 BC (21) Memphis produces WARRIOR

It takes 13 turns to make the settler instead of 11, but you catch up quickly. It only takes 21 turns till you have 2 cities / 2 warriors / 2 workers, and you have a 2nd mine.

By making your settler earlier, you are forced to make your infrastructure later, which puts you behind in the long run.

[barefootbadass]

Ok, I tested this strategy comparing with normal develoment. I build a mine on grassland with shield and then add to the city. In my test games I started on grassland with wheat so my capital city grew quickly, which it seems makes the gambit a bad move.

Based on this, I think Aeson is right about needing a few square or two of game in forest. Most of all, a high food square in your capital city definitely makes it better to keep your worker(because your production can't keep up with the food rate I presume).

Anyway I used the americans, and here is how it went.

Americans, Monarch

Normal development
4000 - Build City(on wheat, whoohoo!), ceremonial burial from hut.
3950 -
3900 -
3850 -
3800 - finish mine on shield grassland, move worker
3750 - finish warrior
3700 -
3650 - city grows to size 2
3600 - finish mine on shield grassland, finish warrior
3550 -
3500 - finish road, move worker
3450 -
3400 -
3350 - finish road, move worker
3300 - finish settler, move settler, found japanese, traded for wheel
3250 - build city

At 3250
2 cities(5 excess food, 6 production, 4 commerce)
1 worker
2 warriors

Now with adding the worker back ...
4000 - build city, ceromonial burial
3950 -
3900
3850
3800 finish mine on shield grassland, move worker
3750 build worker into city
3700
3650 city grows to size 3(note that the warrior is in the city instead of exploring, ML)
3600
3550 get wheel from hut
3500
3450
3400 finish settler, move
3350 move settler
3300 city grows to size 2, build city
3250 build worker

At 3250
2 cities(5 excess food, 5 production, 2 commerce
1 worker
1 warrior

Clearly in this situation you are better off keeping your worker. You can explore better if you can send your first warrior off, which wasn't a big deal with expansionist, but with a non-expansionist civ this might be huge. I'll try it out soon with a map without special food in city radius.

[barefootbadass]

Ok, I used the americans and tested using 2 forests with game in the city radius. Basically the gambit does much better than it did in the previous test and arguably better than not doing the gambit in this scenario.

Americans - Monarch

4000 build city
3950
3900
3850
3800 finish warrior, finish mine
3750
3700 finish road
3650
3600 finish warrior
3550
3500 size 2, finish road(to spices)
3450 finish warrior
3400
3350 finish warrior
3300 finish mine
3250 finish warrior
3200
3150
3100
3050 finish mine
3000 size 3
2950 finish settler
2900 build city(in range of a cow)

two cities(size 1)
5 excess food, 6 production, 3 commerce
1 worker
5 warriors

Adding the worker . . .

4000 build city
3950
3900
3850 finish road
3800 finish warrior
3750
3700
3650 finish road(to spices)
3600 finish warrior
3550 finish wealth
3500 size 2, finish mine, add worker to city
3450
3400
3350
3300 finish settler
3250 build city(in range of a cow)
3200
3150
3100 finish warrior
3050
3000 finish warrior
2950 finish worker
2900 size 2

two cities(size 2 and 1)
5 excess food, 6 production, 3 commerce
1 worker
4 warriors

Now the gambit is only ahead of normal development in population because of the cow, but even without it the second city would be in better shape because the food box would be closer to full. There are fewer terrain improvements but that does not matter much since there are special squares available that do not need them immediately. It is behind by one warrior but in a few turns it would be ahead since it has a current production advantage.

[Aeson]

Thanks for the logs guys, I like testing, but keeping a log isn't one of my favorite things to do, thus most of my observations are rather vague. One thing I was wondering was how commerce was affected in the scenarios? Early on a few extra beakers can mean the difference of several turns of research. Also picking up important techs from the AI with spare gold can mean a huge difference in the long run. Again, just general observation, but I think that even when there is a production/expansion advantage by adding the worker, that the early lack of commerce would offset it.

[barefootbadass]

In the second scenario commerce was about the same either way since there was a little more time for the worker to build improvements before adding back. I built the same 2 squares of road to the site of the second city. They didn't go through the forest with game I was working on in either case because I wanted to get to the spices, but you could have chosen to build through the game to get more commerce. I did maintain my minimal research rate and a paltry +1 or so gold per turn(I usually make money to help buy techs early on by raiding barbarian camps).

In the first scenario there was not enough time(because of high food rate) to get the roads built in addition to the mines, which was another reason it turned out bad.

[Dimension]

I did some more testing with this, and the problem is that in the early game you ALWAYS want to be getting the most possible food for your city.

So, it is a little flawed to say that adding the worker back is more effective when you only have a few forests to work. Basically, you're saying that it's good to add the worker back to the city and make a settler ASAP when your city has low food, i.e. if your first city is on a horrible site, you should abandon it. If the site is that bad, you really shouldn't have founded your first city there.

Forests only give 1 food, so it's ridiculous for a size 1 city to work just a forest. Game has 2 food and 2 shields, so that's nice -- it's as good as a mined grassland (shield), but still, one of the most important things in the early game is to focus on squares that can produce more than 2 food under despotism. If all you have around your first city is game, you should chop down the game forest and irrigate it so it produces 3 food.

[Darth Sidious]

it depends i guess

if you dont have any resources, there might be a slight benefit, but if you have a food bonus, it will be a definite disadvantage.

I am lame enough to restart the game until i have cattle or wheat for my first city, so for me it always sucks. I irrigate the cattle or wheat, and the settler will be ready just as fast.

Also, do i really need the worker because not having it will cost me huge amounts of science.
I prefer to have iron working as fast as possible so that i can get a city at the nearest iron mine asap, so teching is important, and 1 or 2 roads make many turns of difference in tech speed.