Joining Initial Worker to Capital

[Puppeteer]

I've never tried it, but I recall seeing an occasional mention of someone joining their starting worker to the capital city instead of performing worker actions. It occurred to me (too late) in a recent start that I could have had my first settler out at turn 10 if I had done so.

I wouldn't expect that joining the starting worker would usually be a good idea, but I bet there are sometimes it would be a good strategy, particularly in a start with no food bonuses and a couple of BGs or other 2f/1s tiles around.

After a few searches I wasn't able to find discussion of this, but I seem to recall an SG or HOF game where somebody started this way. Does anyone have links to a game or discussion where this was tried?

 

[vmxa]

I would expect it is a tactic one may use on a tiny map playing as Inca. The purpose would be to get the scout out and try to kill your neighbors. You may be aboe to catch them with an undefended capitol.

 

[shl7070]

I think of two possibilities:
1)Extremly bad start, no food bonuses, only one square that can produce 2 food, no irrigation source, pottery not a starting tech, no better terrain in sight.
Then think of building two roads with the worker or road+mine,then join it in the city to produce the settler ASAP and find a normal location.
In almost any case such a start will kill you on the high level anyway.
I don't know what happens if the start is such that city can't grow to pop 3, can you join the worker when the settler shield box is full, the city will be size 3 for one turn, will the settler be produced?

2)The other possibility is the opposite, Cow herd start (5 cows very close) enabling two settler factories on start, use worker to make one road, join him,settler created after only 5 turns of play.
Then establish the second city and then build granary in each city to have two very early settler factories for super REX.

 

[creamcheese]

I think of two possibilities:
Extremly bad start, no food bonuses, only one square that can produce 2 food, no irrigation source, pottery not a starting tech, no better terrain in sight.
Then think of building two roads with the worker or road+mine,then join it in the city to produce the settler ASAP and find a normal location.
In almost any case such a start will kill you on the high level anyway.
I don't know what happens if the start is such that city can't grow to pop 3, can you join the worker when the settler shield box is full, the city will be size 3 for one turn, will the settler be produced?
The other possibility is the opposite, Cow herd start (5 cows very close) enabling two settler factories on start, use worker to make one road, join him,settler created after only 5 turns of play.
Then establish the second city and then build granary in each city to have two very early settler factories for super REX.

I think the city starves down to size 2 before the settler is produced, growth/starvation comes first. With a granary though, you get 1/2 of your food box, and thus can produce a settler. As a practical matter, if I was crazy enough to want to try a start like that I would move the settler instead of settling in place, even 10 turns of moving out of a desert or w/e is preferable to a start that bad.

 

[Calis]

I would expect it is a tactic one may use on a tiny map playing as Inca. The purpose would be to get the scout out and try to kill your neighbors. You may be aboe to catch them with an undefended capitol.

This is the only reason to do it.

 

[Puppeteer]

So getting a second city 10 turns early isn't worth consideration? I'm still thinking it through.

I started a game to compare joining the worker against not joining the worker. My very first try came up with the scenario I had in mind, lucky me.

Emperor Random civs, I got Mongols, other details in spoiler tag:

Spoiler :

• Epic C3C, no mods other than smiley/nation-flag popHeads
• Seed 1496780
• Standard Size
• Roaming Barbs
• Continents 70% Water
• Normal Climate, Temperate, 5 Billion Years
• Random Civ, random AI civs, standard settings minus Culturally linked starts
• Emperor difficulty
• AI Aggression normal
• NoAIPatrol=0

4000 BC Save

The settler is next to fresh water. No food bonus in sight, but 2 BGs are present in the 8 surrounding squares, allowing 2 pop to get 3spt with no tile improvements. The scout will find a forest deer 2 N.

Turn 0, 4000 BC: Found capital on the spot and join the worker, set to build Settler which will finish in 10 turns. Set research to zero (this is just a demonstration) and lux to 20% since this is Emperor and the 2nd pop point is unhappy.

Turn 10: Settler completes, start warrior.

Turn 13: Found Ta-Tu by the forest deer. Since I'm getting 3spt I decide to build a 20s build, in this case an archer, although in a real game I might have built the Mongols' cheap barracks.

Turn 15: Karakorum warrior -> worker.

Turn 20, 3000 BC: Ta-Tu completes archer. Karakorum completes worker.

If I had settled in place and not joined the worker, this is the soonest the settler would have been built, but I have 2 towns, a new worker and 30s worth of other builds at the expense of not improving any land yet.

Totals:

• 20 turns
• 54 food
• 70 shields (1 wasted in production overrun)
• 57 gold earned (had to pay 10g in entertainment expenses at Emperor difficulty)
• consumed starting worker and didn't improve any tiles for 20 turns while there was a deer begging to be chopped and irrigated ASAP
• did not bust any huts with the scout so I can compare fairly to other trials

I still can't find the SG or HOF gamelog where they joined the starting worker. Maybe it was a GOTM....

 

[CKS]

I don't think that anybody in the HOF who joined the starting worker kept a turnlog, because these games are incredibly short. I know that EMan submitted a save for almost every turn of his tiny, fastest conquest games on regent and below.

Here is a link to the chieftain games: http://hof.civfanatics.net/civ3/game_info.php?entryID=1602|1622|562|474|534|464|23|2|376|402

 

[Puppeteer]

Trial 2: Same starting save, settle in place, worker improves tiles

My usual move would be to mine the SW BG, so I could then mine the 2nd BG without roading first, especially if I didn't have an expansionist civ with a scout to see the deer in turn 0.

So I mined the SW BG but decided to pursue the deer next as even a non-expansionist civ would see the deer as soon as it settled in place. I roaded the mined BG before proceeding to the deer because otherwise the deer chop would have beat the founding of the second city.

Turnlog in spoiler tag:

Spoiler :

Turn 0, 4000 BC: Settle in place, Worker SW to BG. 2spt
Turn 1: Worker starts mine
Turn 5: Warrior -> Archer
Turn 7: Worker completed mine, starts road. 3spt
Turn 10: Worker completed road, moves N of city. Karakorum grows to pop 2, 4spt
Turn 11: Worker starts irrigating grass next to deer
Turn 12: Archer (overrun 3s) -> Settler
Turn 15: Worker completes irrigation, starts road.
Turn 18: Worker completes road, moves to forest deer
Turn 19: Worker starts forest chop.
Turn 20, 3000 BC: Settler completes (overrun 2s) (start archer at 3spt)

1 town, 2nd town in 2 turns, still have the starting worker plus a settler, 30s in other builds and tile improvements.

Totals:

• 20 turns
• 40 food
• 60 shields (5 wasted in production overruns)
• 71 gold (none used for entertainment, used an MP)
• Starting worker mined & roaded a BG, irrigated/roaded a grass and started chopping the deer forest, well on the way to setting up a despotic 6-turn settler pump in the second city (not yet founded)
• no huts busted

Okay, joining the starting worker was not the better move in this case. It produced more food and shields in the first 20 turns, but you want that worker back ASAP, and that takes back 20 food and 10 shields, more than wiping out the early-2nd-city apparent food & shield advantage. Also, in this case with the close deer and a worker developing the lands the 2nd city will become a despotic 6-turn settler pump quite early. The commerce production is much closer than I thought it would be. Both Karakorum's spent 10 turns at 2 pop, but trial 1 had 7 turns with a second city and trial 2 had 10 or 11 turns working a roaded tile. Also, although trial 1 built a settler 10 turns earlier, trial 2 had a road so the city was founded only 9 turns after trial 1. It could have been 8 turns if I cared enough to build the road out another tile.

I decided to play a few more turns to see how quickly the settler pump might develop. I might go back and play some more turns on trial 1 or even go back a few turns and start a granary instead of an archer in Ta-Tu on trial 1.

Spoiler :

Turn 22: Found Ta-Tu -> Granary
Turn 23: Worker chops 10s towards Ta-Tu Granary, starts irrigating deer.
Turn 25: Move pop from BG to deer, irrigation in 2 turns for growth in 4 turns
Turn 27: Karakorum Archer (overrun 1s) -> Archer. Worker finishes irrigation, starts road
Turn 29: Ta-Tu grows, need to figure out best MM for completing granary and starting settler pump, but that's out of scope for this trial. (Irrigated deer, 2 forests available, 1 choppable, 2BG, worker ready 1 turn away from stepping onto the workable tiles) Karakorum settler is 11 or 12 turns away depending on whether I want to waste some shields to get the settler out on turn 40.

In turn 29 I need to figure out the optimal combination of city worker placement, growth rate and 2nd forest chop to complete the granary and set up the settler pump. There are too many options, and I'm not playing this for real, so I don't even get into it. Karakorum can produce another settler in turn 40 or 41 depending on whether or not it wants to waste some production to get a settler out 1 turn earlier.

Trial 2 played out to turn 29, 2590 BC:

 

[Puppeteer]

Trial 3: Settle in place, automate worker


Just for fun, trial 3 will settle in place and automate the worker. Heck I'll automate the scout, too. This is the first time I've automated a worker (outside of auto-clean-pollution late game) since the first few games way back when Civ 3 was new.

Turnlog:

Spoiler :

Turn 0, 4000 BC: Found capital -> warrior. Automate worker and set scout to auto-explore.
Automated worker moves E and starts roading BG.
I've been running my scout more or less in the same pattern. The auto-explore scout happened to head straight for the area I haven't looked yet. This is apparently a lonely island start.
Automated worker starts irrigating BG. If you're reading this and automate your workers, realize that the worker is spending 4 turns doing something completely useless for as long I'm in Despotism.
Turn 5: Warrior -> Warrior.
Turn 6 IT: Automated scout pops CB from a hut.
Turn 8: Automated worker finishes irrigation, moves to BG SW of Karakorum.
Turn 9: Automated worker starts roading SW BG.
Turn 10: Karakorum Warrior -> Settler
Turn 13: Automated scout pops 25g from a hut
Turn 16: Notice automated worker irrigated the other BG and has now moved to a forest tile.
Turn 17: Worker continues past forest NE, N of Karakorum.
Turn 18: Worker starts road to nowhere, from nowhere.
Turn 20: Settler completed.

The automated scout did a surprisingly good job; there is only one fogged tile on the whole island, and it's one step away from my borders. The automated worker, on the other hand, was horrible. I am wondering why it went to where it is now. Wild guess: there's a resource there...horses I bet. I can't think of why else it would go there and ignore 2 BGs and 3 forests. (Edit: No, no horse there. I joined the settler back to the capital and researched at max for Wheel. The autoworker mined the lonely roaded grass and then went back to irrigating another BG.)

Totals:

• 40 food
• 50 shields (automated worker left us high and dry...or uselessly wet)
• 86 gold (not counting popped hut or starting gold)
• Automated workers are bad for you
• Automated scout I may have to try a few more times, although this was a smallish island start to the scout's advantage

If you're going to automate your worker, then maybe joining it instead is better. Unless you let the governor run your cities, too.

 

[Ataxerxes]

Joining the initial worker to the capital is almost never a good move. You really need to get the tiles improved ASAP. Trust us, we're not trying to steer you wrong and most of us have a lot of experience. On certain starts you might get your first settler or scout out, but your research will suffer as will your later production and your future settlers. Also, very soon you're going to lose that pop plus a few shields when you do build your first worker.

Workers are the most important unit in the game except for settlers. And on a shield for shield, pop for pop basis they are more important.

 

[Puppeteer]

Well, it seemed that getting a settler out at turn 10 versus turn 20 might be powerful enough to warrant further investigation, but I can see that trial 2 (no worker join) above is a much stronger start than trial 1 even though trial 1 is earning 4fpt at turn 13. (Trial 3 is largely a warning to those who automate workers.)

And I can see now that it would be a very contrived situation (if it exists) where joining the starting worker would be better, and you wouldn't be able to see the need for it in turn 0 even with a scout. Even in the contrived situation and ignoring production and commerce loss from lack of improvements, the lack of roads would slow down the expansion to negate any possible advantage of getting the first settler out earlier.

If nothing else, this thread is a good illustration of the power of the worker.

If I try again, I might move the settler to claim the deer for the capital and see how that trades off against losing the fresh water. (But since I know that this is a lonely island start the "right" move would probably be to settle SE on the coast for growing big and building Colossus after settling the 2nd city to claim the deer.)

 

[Spoonwood]

Trust us, we're not trying to steer you wrong and most of us have a lot of experience.

I think his analysis worth the time, even though most experienced players already know this.

 

[Sparthage]

The automated worker, on the other hand, was horrible. I am wondering why it went to where it is now. Wild guess: there's a resource there...horses I bet. I can't think of why else it would go there and ignore 2 BGs and 3 forests. (Edit: No, no horse there. I joined the settler back to the capital and researched at max for Wheel. The autoworker mined the lonely roaded grass and then went back to irrigating another BG.)

I wonder if there really is a resource there, and if so which one it is. Nice analysis though. I don't think I've ever tried to join my first worker.

 

[Raliuven]

Thanks for taking the time to do the analysis. Confirm or deny, its all good information.

 

[GamezRule]

If you can post the seed number or a save, you need a third party site atm, we can see if there is a resource there or not.

 

[Puppeteer]

If you can post the seed number or a save, you need a third party site atm, we can see if there is a resource there or not.

Both seed and save are in post#6 just above the starting location image.

And here:

Seed 1496780
4000 BC

 

[Hellfiredoom]

Thanks for taking the time to do the analysis. Confirm or deny, its all good information.

I enjoy these reads. Most of my games follow the same generic route because I use the same Civ and starting rules each game. Not complaining, just stating. So it's fun to read when someone tries something "outside the box".