Ducki Does C3C at Emperor

[Dominae]

The demi2 game in the other thread supports a couple of pumps. They're a bit harder to exploit, but that's part of the fun.

[cracky]

MMmmmm! beef!

I'll give this one a shot.

I've been making a few feeble attempts at Emperor after winning a wee bit too handily on monarch recently (this *is* an Emperor game, no?). The research pace is still a bit frenetic for me, but interesting, no doubt. I'll post some screenshots along the way and any advice would be greatly appreciated!

Just to double check, the idea here in your normal early game, REX, etc. and once appropriate to work on this population transfer, these forced worker migrations to new, *better* environs!

[Aqualung71]

Originally posted by ducki
.....the concept of maintaining a worker pump for the express purpose of transporting cheap labor from one town to another where he is inherently two, four, six, maybe eight times more valuable(food cost, immediate commerce/production).

In recent games I have started adding workers to towns during that unproductive period (for workers, that is) between ~ mid-middle ages when you have all your usable tiles roaded and improved, and Steampower when you need them for RR'ing. Dropping 4 workers into a 2-pop coastal town with a Harbour adds 12 commerce before corruption. Do this several times in a short space of time and you've significantly increased your income and research potential. As long as you keep your camps pumping out workers (or military units when you've reached your infrastructure-imposed pop limits and don't need more workers), you will be able to quickly build up again for the RR's.

This has made me look at workers differently - more in terms of what it is costing me to keep them there, both in the unit upkeep and the commerce lost from them not working a tile as a city labourer. And in that regard, if you're in Republic and you add a worker to bring a town to size 7, you gain even more because your unit support cost goes down (stock rules).

I can see how some would argue that building workers then adding them back to a town/city is a zero sum gain, except that it wastes shields that could have been spent building something else - and the justification is that you used them to improve worked tiles, which in turn creates more shields/food/commerce. But this misses a crucial point: when you build the worker in a town and add it to a city/metro, from a growth perspective you have effectively reduced the food box in all your cities/metros to town size levels (ie 20, or 10 with a granary), allowing faster growth of your empire overall. This is a huge growth lever that I think is largely overlooked.

Once all RR's are built and land terra-formed (ie, irrigation switched to mining), disband your camps and add the remaining workers to your selected mega-cities (with hospitals) to work the tiles previously worked by the camps. And now you've got a few mega-production metro's ready for the SS build. (This would work a charm, if the AI didn't keep declaring war and screwing up all your neat plans)

As Dom said, the game Lethe posted provided the opportunity for this, and that is where I learnt to do it (it was actually the demigod.sav about 4 posts down on that thread that gave the cow start). The number of AI civ's is lower than normal on this game, so it gives you plenty of space to expand and practice these techniques.

[cracky]

Well, I *was* going to bed, but...

Ok. I played until 2070 BC. I have the 4 turn settler pump set up. Kyoto's build queue was warrior, settler, worker, granary, settler, settler, worker, to be followed by settlers, yea until we have been fruitful and multiplied across the land. I needed the second built worker (3rd worker in total, if that makes sense) from my settler city because my worker pump ain't quite right in the head yet. At the moment, it still has 57 turns on a granary (to be shortened by wood chopping of course).

Osaka was settled 1112 of Kyoto, just north of that mountain. After my initial two workers from Kyoto set up the pump, one went to work building a road straight up a mountain into a diamond mine, and as of this turn, it still isn't finished! Osaka has built a warrior, a worker, and a settler so far (which I guess is why it's so far behind on the worker pump granary).

mini-spoilers to follow...


I have 4 cities with a settler on the way to securing the dyes. Right after Kyoto builds a worker, the next settler is going to secure me some ivory. HOWEVER, I have only three workers at the moment (bad Cracky, bad bad Cracky!). Will have a 4th in 2 turns and a fifth in 4-5 turns. The general plan right now is to spread like locusts while Tokyo (22222 from Kyoto) builds an MP, a barracks and many archers. Edo (2333 from Kyoto) might do the same with a settler or worker thrown into the mix as needed. I'm not really a WarMonger (stop snickering!), but Russia is going to be seriously cramping my style in the not too distant future.

I was planning on attaching a screenshot, but I managed to screw it up somehow and reconnecting my Civ3 PC to the net to try again seems like a bit too much of a hassle at the moment. I do have a save though. Again, any and all comments would be great!


[edit...]

Duh, forgot the part about the worker pump!

I've never really played a game with camps, so there's a good chance I'm way off here...
886 of Kyoto seems to be a great place for a worker pump. No improvements, nothing but shields and food for workers. ...but I still think it's more important to finish the land grab first...

[Arrian]

That site is FREAKING AWESOME, ducki. It can DEFINITELY be a 4-turn pump. In fact, if you are so inclined, I think it can be turned into TWO pumps (one for setters, one for workers). The ease of that second option depends a bit on what the land to the southwest looks like. EDIT: having re-read your original post, I realize now you saw that potential too.

Option1: 1 pump, built where you stand.

Mine both grassland cows (each 3f, 2s).
Mine grassland wheat (3f, 1s).
Mine 3 bonus grasslands (2f, 2s each), of which you actually have 4 available.

At size 5, the city is using the 2 cows, the wheat and 2 bgrass. This gets it to +5fpt, +10spt. It's actually so shield-rich that you can build THREE turn settlers (though the growth can't keep up with that). So, if you can squeeze in a barracks (you can, I'm sure), you can either go settler/warrior/settler/warrior in preparation for a sword upgrade, or you could go settler/worker/spear/worker/settler.

Option 2 coming.

-Arrian

[Arrian]

Option2: Two pumps.

Mine the cow across the river, but irrigate the other. Boom, right there you're at +5fpt. The city can use 2 mined bgrass and 1 normal mined grass, plus those two cows, for +5fpt, +9spt at size 5. Hmm, especially with that forest tile available, that city is STILL capable of the stuff I mentioned above (it will average 10spt or more, meaning it will produce settlers too fast to just build them back-to-back-to-back, so you can mix in other stuff).

Meanwhile, you build a city 111 of Kyoto. Irrigate the wheat tile. That right there gets the city to +4fpt. Now, it gets trickey if that city site does not have any other bonus food available. Let's assume it doesn't. In order to get this one going as a 2-turn worker pump, we can go back and irrigate over the 2nd cow at Kyoto3 (this gets Kyoto to +6fpt/+8spt at size5). Now we start sharing the other irrigated cow between Kyoto and Osaka.

One one turn, Kyoto uses both cows, getting +6food. Osaka, meanwhile, uses its wheat and whatever other tiles its got for +4food. The next turn, Kyoto hands off the cow to Osaka, reversing the food situation so that Kyoto gets +4 food (total of 10, and growth) and Osaka gets +6 food (again, 10 total, and growth).

Booyah, twin pumps. REXplosion.

-Arrian

[ducki]

But this misses a crucial point: when you build the worker in a town and add it to a city/metro, from a growth perspective you have effectively reduced the food box in all your cities/metros to town size levels (ie 20, or 10 with a granary), allowing faster growth of your empire overall. This is a huge growth lever that I think is largely overlooked.

This is precisely the bit that I didn't notice being explained - taking a 10f10s worker from 1 town and magically transforming it into a 20, 40, 60 food laborer in another town - with a 2 turn growth/production phase and then however long it takes to get where he's going - saving probably half the growth time in a metro, maybe more.

(Minor spoilage follows)
@Arrian - I built the Settler Pump on the spot, irrigating both Grassland Cows and wheat(is that in the screenie?), mining all the BGs AND Gs, sharing a Cow between the capitol and the Worker pump settled 111(I think) from the capitol. I knew for sure I could get the settler pump, I still haven't finished mining the second town to ensure I have enough shields for the Worker pump.

Also, Russia is exceedingly close in this one, so much that I "had" to build a temple in the Worker pump because it was settled a mere 1 tile away from a Russian town(C-x-C). Actually, I had to squeeze out the settler before the granary to be sure I actually GOT the second pump site I wanted(river, shared cow, inland/toward AI from capitol, horses. The lack of extra forest slowed this one down, as I didn't want to spend time mining a hill or mountain and therefore couldn't afford to chop the forest.

Since I'm looking to explore the "build workers for pop-transport" idea, I don't think mining any bonus will allow two pumps, but I'm still learning this, I may have missed something. If I only wanted one pump, I still might irrigate everything just so I could still share the bonus food between my towns since I still try for a CxxC spacing in the beginning - no sense wasting tiles, and a good thing with Russia so close.

Looking forward to your version 2, maybe it will have less MM/tilesharing than mine.
Edit: Crossposted with you Arrian. Glad to see I didn't miss anything and all that tile-sharing/switching wasn't overdoing it.

[Arrian]

You did option2, exactly, it seems.

OUCH, them Russkies be close! Is their start location really close too, or is this a city/settler from hut deal? Or just bizzaro AI settlement in Conquests?

Anyway, my answer wouldn't have been a temple, I don't think. Every city other than the 2 pumps would be ordered to build barracks right away and then pump troops. Any civ that close must die.

-Arrian

[ducki]

I think the spacing was a bit close in this start - what can ya do, it's a RNG thing. Seems to be J-c-c-R on my map, J=my capitol, R=Russias, c=city spot. Mighty tight.

Yeah, I probably should have just built several warriors instead of the temple, but it was so cheap and it will help keep my lux-tax down anyway if I'm shield-poor enough to need to do the worker pump at size 5 instead of 4. Also, that tight, I got a little worried about actually having spots for my future settlers, maybe a bit of border jump will keep the AI at bay a tad - that might have been my logic, partially.

But yeah, churning out some chariots instead would probably have been a better use of the shields, I've just never been squeezed that early and kinda freaked out, fell back on my old builderliness.

[Arrian]

Yeah, I probably should have just built several warriors instead of the temple, but it was so cheap and it will help keep my lux-tax down anyway if I'm shield-poor enough to need to do the worker pump at size 5 instead of 4. Also, that tight, I got a little worried about actually having spots for my future settlers, maybe a bit of border jump will keep the AI at bay a tad - that might have been my logic, partially.

Doh, forgot the Japanese were religious! In that case, the temple makes perfect sense. 30 shields is worth it. 60, otoh, is another matter.

-Arrian

[Tolwyn]

okay. I've been trying the spanish map from the start of this thread. I understand the theory behind the settler-pump but I just can't seem to execute it in practice.
Can someone pm me a play-by-play? In the meanwhile I'm going to bed. My head hurts in pondering why I can't figure this out

[vmxa1]

IIRC there is a good example in the AU501 thread before the game was even started.

[ducki]

Originally posted by vmxa1
IIRC there is a good example in the AU501 thread before the game was even started.

There is indeed. It was in the spoiler thread and had quite a bit of Turn-by-turn analysis. It got quite spooky, actually.

Here is the thread and here is a link to Arrian's Turn-by-Turn, complete with worker actions, running shield and food counts, and build orders.
And
here is Alexman's pure expansion "farmer's" gambit zero military opening.

Hope that helps!

[Tolwyn]

It sure does.
/me bows in humility

[Harovan]

Originally posted by Arrian
Doh, forgot the Japanese were religious! In that case, the temple makes perfect sense. 30 shields is worth it. 60, otoh, is another matter.

-Arrian

What, YOU forgot that the Japanese are religious? Who are you, and what have you done with Arrian? Thanks, that saved my day, after the announcement of the weak 1.20 patch ruined it.