Why growing to 5 is bad OR the value or a warrior

One bit of advice commonly doled out is grow to 5 and pump workers/settlers.

I think it's bad and inefficient. Here's why:

THE COST
Growing from size 4 to 5: 28 food.
1 defender (cheapest is a warrior): 15 hammers. *unless you have ivory/gold or are charismatic
That's a total of 43 food/hammers. That's almost half a naked settler, or more if you're imperialistic. Or 3/4 of a worker.


THE BENEFIT
Most likely your 4th-5th tiles that you work are going to be at best 4 yield tiles, like mines or farmed flood plains. So you're gaining 2 extra hammers+hammers.

THE OPPORTUNITY COST COMPARISON
In a heavily forested area, a worker can chop 20 hammers in 4 turns for 5 hammers/turn. 43/60 of that is a little under 4 hammers/turn. Later he can go to your other cities.
A settler with an escort is 115-125 hammers. A new city working a 5 yield improvement produces 6 yield. 1/3 of that is 2 hammers a turn. Of course it can grow and work an additional mine for 8 yield.

Warriors:
Your warriors/archers are valuable early. I tend to make around 2 growing to size 4, sometimes less if I have a high food start. If you have a low food start (plains cows+hills), you're probably going to have too many warriors.
1st warrior, your starting one, scouts but often dies to a bad roll from a lion or bear.
2nd warrior scouts/escorts.
3rd warrior scouts/escorts/fogbusts and possibly escorts your 3rd settler.
Garrisoning them does none of this. Barbarians won't even attack your capital for a certain period of time, so it's wasting time.

That 5th population is much more useful when you cannot expand any further or your capital's garrison becomes relevant. This is usually either due to lack of land or lack of commerce.

Is there a counterargument for this

When the 5th tile is a plains / grass hill you can often convert that last tile to +4 with imp when producing settlers and +3 with exp when producing workers. Yes the opporunity cost can be large, but as you said you quite often have spare warriors anyways and hence growing to size 5 doesn't cost you 43, but 28(you just have one less explorer/fogbuster and one MP in the capital). Often you don't have a heavily forested area, and even if you do chopping all those forests asap is not THAT much better than chopping them sligthly later(or maybe even chop them with math).

Counteragurment for me is city specialization. Your capital is going to build say 4 settlers and 6 workers before anything else. Your others cities are going to build a monument, granary, maybe barracks before they could build worker / settler. So the yield of that cannot be added to the equation until much later. And eventually you will grow it to 5 so might as well take the benefits from it. That being said, if I need to grab a strategic resource or a floodplain / gold city I will postpone the growth.

Second argument is I want to scout out my territory before I decide on city locations but I am probably to anal about that because the AI usually ruins half of my plans by building an inconveniant city close to my target locations.

There was a thread somewhere that proved size 4 was the optimal size for building settlers. Usally i'll always grow to 4 before a starting settler. with everything it is map dependant.

I use 4 myself, but I will go to 5 with CHA leaders, on my way to 6. The 3-pop whip with three 5-6F food sources is pretty sweet.

I typically build my first settler at size 3 or 4, but it depends on the tiles I've got. If I have a lot of seafood, for instance, I tend to get that online before the settler, which is a big delay and upfront hammer investment. So city 2 is slower to appear, but I have better worker/settler production (not counting chop/pop rushes) and better research.

Other times you just have two visible workable bonus tiles, and you need to get that first settler chopped out faster. I will build a settler at size 2 under those circumstances.

And then there's the situation where you have some floodplains to start, and another floodplain city to settle. Arguably settler first is an option there, since city 2 will get founded and growing much sooner, even working unimproved tiles. I very rarely do this, but I've seen it done here on CFC.

myself, i like to grow to five. it makes building axes so much quicker

fjordan said:

So the yield of that cannot be added to the equation until much later. And eventually you will grow it to 5 so might as well take the benefits from it.

Click to expand...

This is not a good way of looking at the situation if you're trying to optimize. You're going to get a granary, too, but building it first tends to delay your first city substantially.

Concerning delayed returns: my argument is transitivity. With a newly settled city, you could choose to go straight for another settler. However, growing to size 2 with 2 special improvements has greater yield than building a straight settler. So by transitivity, growing in second city > building worker/settler in 2nd city > growing to size 5 in first city. It's a greedy algorithm, even if it's delayed, it's still a more efficient strategy than growing.

If you're worried about future yield return, instead of looking at total food/hammers output in your empire (which I find to be the best measure), you could look at the number of turns it takes to recoup your initial investment. For example, using the 43 food/hammers figure, it would take 21.5 turns to recover that investment. A settler takes 25 turns + moving time to return its investment with no improvements. With say a 5 yield farm improvement, it's 5+(100-4*5)/6=18 turns + moving.
But again, this is a worse model than total yield, since REXing approaches exponential growth, so your base term is the most important factor.

I find that game situations seldom yield this being a purely mathematical decision. I'll whip a settler before a 2nd worker if I really, really want a city spot the AI is going to get otherwise and I don't think I can/should rush it. Having all the time in the world or a need for fogbust troops changes it too.

Of course, if you really have huge food/low hammer starts then there's really no issue ---> you're going to be whipping, and you're going to be capped at 4 until you get more sources. Warriors are probably just overflowed in this scenario.

"It depends". You wanna get that food going, which means at least 1 preferably 2 workers, or equivalent workboats. After that you can either grow if city 2 can wait (for example, you haven't gotten to BW yet or haven't scouted enough yet to see where copper might be hiding), or get the settler building on the high food. Whip to taste.

In general, I wouldn't criticize your strategy, but it really, really does depend. If you're food-heavy enough, it can be worth the time to grow.

And you have to look at what you're going to be building. Do you have a second city site that is practically a second capital or are you in the middle of a jungle? Did you steal a couple workers that have nothing important to do anyway or do you desperately need another worker ASAP?

Growing to FIVE!?

Growing to 2 before worker is bad.

Growing to 2 before settler it depends on whether you have 2 good resource tiles that you can improve both of.
Growing to 3 before settler is bad.

Alexfrog said:

Growing to FIVE!?

Growing to 2 before worker is bad.

Growing to 2 before settler it depends on whether you have 2 good resource tiles that you can improve both of.
Growing to 3 before settler is bad.

Click to expand...

I'm pretty sure he's assuming a worker 1st build. Even worker 1st isn't always optimal but it is very, very frequently the case that you want to open worker 1st (unless you're playing a difficulty you can warrior rush ---> warrior x5 is cheaper than a worker and settler and will often result in you getting both)

Growing to e before settler is not bad. In fact, it's very frequently optimal.

Now I chose only Khmer, Exp&Cre I build one worker, chop two more, and in some cases(enough forest and nothing to work on) I may chop the fourth and than, my first warrior.

of cause I keep my city growing while chopping workers. in most cases it may reach POP 3, and then whip a cheap granny or chop my settler

IMO, the game's focus is on Food resource&Forest. F&F means brutal power...

If I'm in bad luck and have no one to kill with my F&F army... build world wonders and cottage like hell.

Alexfrog said:

Growing to FIVE!?

Growing to 2 before worker is bad.

Growing to 2 before settler it depends on whether you have 2 good resource tiles that you can improve both of.
Growing to 3 before settler is bad.

Click to expand...

The idea is to turn the capital into a worker/settler pump at size 5. You will typically have two workers trained before reaching that point. Next build is supposed to be a settler. I follow this rule most of the time. For example, in one recent game my build order went:

worker
switch-to-workboat
back-to-worker
warrior
warrior
worker (at size three)
warrior
warrior
settler (at size five)

Once you reach size 5 you build nothing but workers and settlers for a long time. The other cities provide military and the like.

Vicawoo is arguing that size 4 is enough before a settler and I agree that sometimes it is. The number of times where it is correct to train a settler at size 2 is extremely rare. You need to get your capital improved. You also need explorers and escorts.

Alexfrog said:

Growing to FIVE!?

Growing to 2 before worker is bad.

Click to expand...

Not if you can finish bronze working within 5-6 turns of growing to pop 2. That's 5-6 turns of working an extra tile before you whip out that first worker at around the same time it would have built manually anyway.
For any civ starting with mining and a floodplain or food resource it's almost always optimal to grow to two and whip out the first worker.

For the other civs is still often the optimal strategy.

vicawoo said:

Most likely your 4th-5th tiles that you work are going to be at best 4 yield tiles, like mines or farmed flood plains.

Click to expand...

This is not always true.

You should keep growing your capital until you run out of 5 yield resources.

TheMeInTeam said:

I find that game situations seldom yield this being a purely mathematical decision. I'll whip a settler before a 2nd worker if I really, really want a city spot the AI is going to get otherwise and I don't think I can/should rush it. Having all the time in the world or a need for fogbust troops changes it too.

Of course, if you really have huge food/low hammer starts then there's really no issue ---> you're going to be whipping, and you're going to be capped at 4 until you get more sources. Warriors are probably just overflowed in this scenario.

Click to expand...

Choosing a suboptimal path is due to non-ideal conditions and additional constraints (such as rushing for land). However, seeing how growing to 5 is slower and suboptimal, not having to rush for land should not cause you to do that.

Or more clearly, growing to 5 < growing to 4, assuming ideal conditions.
Rushing for land means you grow earlier.
Growing to 5 means you grow later.
Rushing for land does not mean you grow to size 5.

Fogbusting requires an extra warrior, so you should be less inclined to keep a warrior at your capital.
I'm not inclined to whip with high food starts pre-granaries unless: I need a settler, or I need to build an extra warrior/axe while regrowing.

The best reason to grow to 5 is if you can't safely expand at size 4.

In general, I wouldn't criticize your strategy, but it really, really does depend. If you're food-heavy enough, it can be worth the time to grow.

And you have to look at what you're going to be building. Do you have a second city site that is practically a second capital or are you in the middle of a jungle? Did you steal a couple workers that have nothing important to do anyway or do you desperately need another worker ASAP?

Click to expand...

If you're food heavy, you're also producing workers/settlers faster. If you're low food higher hammers, it takes longer to grow but your workers/settlers are slower. The ratios between growing and worker production do not change. It does mean that you'll produce fewer warriors while growing, makes that garrison even harder to justify.

In more qualitative terms:

If you're food heavy, you can't spare warriors.
If you're hammer heavy, you'll produce too many warriors and probably should consider expanding earlier (maybe size 2 or 3).

DaveMcW said:

This is not always true.

You should keep growing your capital until you run out of 5 yield resources.

Click to expand...

I'd say this is true in at least 80% (I'd like to say 90-95%) of starts, heavy seafood starts not withstanding.

In the case of seafood starts, you have to build a lighthouse to get to 5 yield, and probably 3 workboats (150 hammers, 120 as organized) and the warrior. I'm not saying the commerce isn't worth it, but I'm very suspicious of such high hammer costs.

DaveMcW said:

This is not always true.

You should keep growing your capital until you run out of 5 yield resources.

Click to expand...

I replayed the opening I described above, attempting follow your rule. There are some errors in the way I described it. Here's how it actually worked. It's from the latest Emperor Cookbook, BTW, so if you want to try the start feel free. I moved one north from the starting position, reasoning that the Blue Circle had probably spotted seafood. Sure enough it did. If you replay it, please do the same.

The capital was built on a plains hill and had a pigs, a fish, a forested deer and plains stone in the BFC. We start with TW and Hunting. Not the best starting techs but certainly useful given the position.

I started a worker while researching Fishing. Next came AH, Mining and Masonry. Right? Work those five-resource tiles. When Fishing came in, I switched to a workboat. Back to the worker who finished just in time to start corralling the Pigs. Road, then camp and road the Deer. Move on to the Stone.

Meanwhile my scout died early to a lion, defending on a forest hill Build continued with two warriors. By this time, the capital had reached size four. The first warrior went out to explore. The other looked around a bit and then came back to fogbust at the selected site of my second city.

Next I trained a second worker. Even though I still didn't have Masonry, the two workers could build a few roads and then work together on the Quarry.

After I finished the second worker, I switched to a third warrior. After all, the capital needs an MP. I discovered that the warrior would arrive in two turns with growth in three, so I switched to working the lake. This meant that both the warrior and growth would arrive in three, giving me some extra commerce on top of it.

Result: I arrived at size five and my fifth tile was only a three resource (with potential to become four once I learned BW, which was my next tech).

Is there something wrong with this sequence? Should I have only trained two warriors? That means that either I would have had no explorer or no MP in the capital. The latter is possible I suppose. Surely you are not going to send the settler out on its own. I should add that, after the settler, my next build in the capital was a workboat for the second city. I couldn't have done that without the MP.

This is my general experience. I arrive at size five before I am ready for my first settler. Not always, obviously. But usually.

Possibly, it might have been better to finish the sequence with settler-warrior-workboat instead of warrior-settler-workboat. I'm not convinced though, because the second city is not valuable until it has its food. In any case, the difference between the two is very small.