Production Plan Proposals

[ZargonX]

Ok, I've been looking things over, and I've put together a few proposals for some more long-term planning in terms of production and worker actions.

Legopolis
--------------------
-The three options, as I see it, for upcoming production in Legopolis involves three choices.
1. If we want Legopolis to continue functioning as a settler farm, investing in a granary wouldn't be such a bad idea. While it will slow settler output temporarily, it will benefit this function in the long run.
2. If we don't want a pure settler center, we should at least build one more settler after the next Merc, and follow up with either a temple to establish an early culture-core, or more Mercs to supply new cities with protection.
3. After the merc, we could also consider producing another worker in Legopolis. This could be used first and formorst to link the road network to Panama, and then subsequently to help Panama get off the ground. Indeed, we could even consider creating two workers, assigning the second to begin the arduous task of clearing the nearby jungles, but that is a task that can be delayed in lieu of the other options.

Jackson
------------------
Following the production of the next settler in Jackson, which seems to be the choice of the majority, we have the issue as to whether to make it a wonder production center. This has been thoroughly discussed in the Jackson Production thread, so I will not rehash it here.
However, as for Worker1, upon completion of his current mining duties, he should move to the sea-cow and commence mining there (2190 BC, I think...). Following that, if we indeed want Jackson as the wonder-builder, mining of the nearby hills would be the next task.

Farmerville
----------------------
As has already been ordered, a rush-built warrior will start things off and hold off the barbarian hordes. Following this, I feel we should build a worker, who can then irrigate the nearby wheat to set up Farmerville from its obvious task: settler farm. Once the wheat is irrigated, begin mining the two bonus squares for a production boost, and then start the city cranking out settlers as fast as possible. With the city running at full speed, we should be able to settle the southern lands with relative expedience.

So, there you have it. As with all plans, these are of course subject to change with external factors. I would like to hear everyone's opinions, so hopefully we can have a good solid framework in place to work with.

[Tiberius]

My choices would be:

Legopolis
- mercenary (the current one)
- worker (we need the worker quickly, because with a granary it will grow quickly and doesn't have enough worked tiles; plus we need the connections to Panama and RedBricks)
- granary
- worker (if needed)
- next: mercenary+settler pairs, and a temple at some point, in between 2 pairs.

Jackson
- settler (the current one)
- mercenary
- wonder (palace)
- wonder (palace)
- temple
The worker, after finishing his current job, should mine the cattle, the road and mine the 3rd binus grass, the mine and road the 2 southern hills.

Farmerville
- warrior (the current one)
- worker
- settler
- settler
- settler
I'd road and mine mine first the bonus grasslands, as Farmerville is growing very fast even without the irrigated wheat (actually it will grow faster than produce settlers). Once we will have irrigated the wheat, we will have to mine a desert tile for more shields.

Panama
- mercenary
- worker (to quickly irrigate the plains and build some roads and mines)
- mercenary
- galley
- galley
- mercenary
- settler
- possibly a temple for culture and one more plain
- wonder (maybe)

[vondrack]

As for me:

Legopolis: finish Merc, then granary, then worker
Legopolis: finish Merc, then worker, then granary
EDIT: some math Tibi provoked showed it would be better to build the worker first - see few posts later
Jackson: finish Settler, then Merc, then wonder
Panama: Merc, Merc/Galley, Galley (no workers/settlers - the city grows too slowly)
Farmerville: rush Warrior, then worker

I do not dare to plan beyond this... and all of this is of course barring any unexpected developments.

EDIT: I would go for a granary in Legopolis first, as Legopolis is currently approaching the optimal population for such a build (3-4). If we build the granary first, worker next, we will be done quicker than if we do the worker first and granary next. Although I know we (will) need workers, Panama will not need any for its first ten turns (working the unimproved game forest anyway), and later the worker currently improving Jackson will be able to move into the area and irrigate the riverbank plains either in time or with a small delay. Proved to be arguable. Building a worker first actually delays getting the whole bundle by two turns only - while getting us 12 worker-able turns.

[Tiberius]

Radek do you think we can build a galley so soon?

[vondrack]

Originally posted by Tiberius
Radek do you think we can build a galley so soon?

That "Merc/Galley" entry means we start a merc and if there is a chance, we switch to a galley later. Writing is now 5 turns away, Map Making shall be something like 16 (if I remember your figures from the other thread correctly). Panama will need 10 turns to build the first Merc... the second one will take less, but our research might speed up a bit as well... dunno... did not do any special math, just indicated that if there is a possibility to switch to a galley, we shall use it. If not, two mercs first, then a galley in Panama.

EDIT: considering the waste in Panama, the first Merc will take even longer... maybe as much as 13-15 turns.

[Tiberius]

Originally posted by vondrack
I would go for a granary in Legopolis first, as Legopolis is currently approaching the optimal population for such a build (3-4). If we build the granary first, worker next, we will be done quicker than if we do the worker first and granary next.

Interesting. You might be right. However, by the time the granary will have been built, Legopolis will be using 4 or 5 tiles, but unefficiently because it has only 3 worked tiles. Hence my plan to build the worker first.

Although I know we (will) need workers, Panama will not need any for its first ten turns (working the unimproved game forest anyway), and later the worker currently improving Jackson will be able to move into the area and irrigate the riverbank plains either in time or with a small delay.

Radek, the worker that currently improves Jackson won't go to Panama for a very-very long time. It has to build a mine on the cattle, road+mine the last grass and road+mine 2 hills. These are a lot of turns. Panama will need a worker from elsewhere; it must be connected with Legopolis, needs an irrigation ASAP, plus some mines on the hills.

Edit: I'll try to run a sim to see the exact figures and compare some possibilities (like granary+worker or worker+granary). It is pointless to argue until we know more exact data

[vondrack]

Originally posted by Tiberius
Interesting. You might be right. However, by the time the granary will have been built, Legopolis will be using 4 or 5 tiles, but unefficiently because it has only 3 worked tiles. Hence my plan to build the worker first.

OK, I did not do any math before... so let's do some:

Granary (60 shields) - Worker (10 shields):
4 turns to finish the Merc. Then granary: 3 turns at 5 shields (= 15 shields), 45 to go. New shield output (at pop 4) will be 7, means 7 more turns to build a granary. Then just 2 more turns to build a worker. Total time: 4+3+7+2=16 turns.

Worker (10 shields) - Granary (60 shields):
The other way round: 4 turns to finish the Merc. 2 turns at 5 shields and a worker is born. Now Granary: 1 turn at 3 shields (3), 7 turns at 5 shields (35), 4 turns at 7 shields (28) to finish. Total time: 4+2+1+7+4=18 turns.

I'm not sure if the granary would be effective in time to save the food for Legopolis while entering pop 4 in the first case, but even if it did... I think it's quite obvious now that it is better to make a worker first - the work he will do over those 12 turns is worth more than two saved turns...). I hereby change my mind and vote for finishing the Merc, then building a worker, and only then a granary in Legopolis.

Originally posted by Tiberius
Radek, the worker that currently improves Jackson won't go to Panama for a very-very long time. It has to build a mine on the cattle, road+mine the last grass and road+mine 2 hills. These are a lot of turns. Panama will need a worker from elsewhere; it must be connected with Legopolis, needs an irrigation ASAP, plus some mines on the hills.

Well, this is now irrelevant anyway, but Jackson will drop to pop 1-2 after spitting out the settler and will not need that many improved tiles for quite some time (as it grows once every 7 turns only). That was my idea. Anyway, let's go for the worker in Legopolis first...

EDIT: Oops, made a mistake which changed the exact result a bit... but not the final decision.

[Tiberius]

OK, I've run a quite detailed simulation, here are some results. I made some mistakes (like I forgot about Farmville and its unhappy citizens and left them alone without military police) and I will have to replay it, so I'm not going to post now the whole table I made, but for the rest of it it's correct, IMHO.

So:
1. Jackson: with the settler - mercenary - wonder/palace build order, Jackson will finish the Lighthouse in 885 BC (in 49 turns from now).
2. Legopolis: in the same period, Legopolis was able to build the following items: mercenary, worker, granary, mercenary, settler, mercenary, settler, temple (in 45 turns)
3. Panama: mercenary, worker, galley, galley, mercenary
4. Farmerville: worker, settler, settler, settler (I roaded and mined two bonus grass, roaded + mined a desert and roaded + irrigated the wheat; it never grew bigger than 4 and never shrank smaller than 2)

Some interesting observations:
- the worker near Jackson finished all his work near the city only after 32 turns!! and after that built a road on the nearest dye because Legopolis extended his cultural borders. I moved him then to help cut down the jungle near Red Bricks. We will need definitely some other worker for Panama (from Panama itself or from Legopolis).
- one more thing: the workers near Legopolis and Jackson were busy all the time, making improvements just before the cities grew. For example, in 1560BC both Legoand Jackson grew to 4 while the workers were improving the 5th tile respectively. I moved then the worker from Legopolis to improve Panama. (but I did build a worker there earlier that in the meantime had built an irrigation, 3 roads and a mine on a riverhill).

[Sharpe]

A fair amount of what I am reading sounds good, except for one thing - we wouldn't be building enough settlers!!

In Tiberius's last list we would be building 6 settlers in 45 turns and half of those would come from Farmerville.

So far, Vondrack's colour map indicated these possible sites:

RedBricks (to be built with Jackson's settler)
Forkmouth
Sharpehaven
Kloreepville (admittedly won't be viable for a while)
DyeFields
and 2 to 3 more spots that were more for filling up the map near the northern mountains, on the isthmus and in the jungle.

True, some of these new cities will eventually be able to start producing settlers, but only near the end of the 45 turns if even that.

After Vondrack posted that map, Gaul and Conan have uncovered even more potential city sites in the south. Admittedly we would start to run into corruption problems down there until we completed the Forbidden City wonder.

Is there any way that we can increase the number of settlers produced in the next 45 to 50 turns from 6 to say 8 or even 9?

The only way I see is to have a second settler city like Farmerville or have two additional cities that concentrate more on building settlers.

Panama is not a good choice as it won't grow quick enough to build many settlers.

RedBricks will have severe shield problems especially early on due to the jungle and its grassland non-shielded squares.

We have strongly considered using Jackson to start building wonders.

Basically, that just leaves Legopolis unless we want to delay building wonders at Jackson.

Long-term Forkmouth and Sharpeville (especially when its grasslands are mined) will be able to join Farmerville as settler cities and shift the settler burden from the northern cities and to be able to reach the southern cow cities. However, it would take 15 to 20 turns after their founding for this to start happening.

My suggestion would be to plan a building strategy around the settlers and slot other items around them. If that means building an archer or warrior rather than a merc in a city so be it. We can always upgrade the warriors to swordsmen later.

[Tiberius]

Sharpe, Legopolis could build another settler after the temple in a few turns, so that would make 7 settlers in 50 turns.
From the new cities, Red Brick should build 2 wokers and then maybe a mercenary or archer, but not settler, until we clear some jungle.
I think Forkmouth could build a settler, but don't know exactly when.
Other cities wouldn't have the time or the resources, IMO.

That makes 8 settlers in 50 turns. Is it that bad?

I'll try today another sim to see if we can build more settlers. Btw, the 2310 legosim should be available soon, as I sent the file for hosting to Vondrack and Kloreep. Maybe others are willing to run the sim, too

[Sharpe]

That makes 8 settlers in 50 turns. Is it that bad?

Nope not too bad - certainly an enhancement on 6 settlers in 45 turns.

If we are alone, this gives us a longer time to do REXing, so we should take advantage of that as well as trying to contact the other civs.

As it is, we may currently have a 1 city lead on the others (but perhaps not as we are consistency 2nd in population).

[Tiberius]

ND are expansionists (arabs) and maybe they've got a town or settler from a goody hut. Having the same number of points as we have, there's a high probability that they have the same number of cities and probably they are no.1 in population.

[Tiberius]

I've run a sim to see whether a granary will pay out on the long run or not (and if yes, how fast) ?

Let's see:

Turn	Year	Legopolis 1	Legopolis 2
1	2310BC	-	-
4	2190BC	builds merc.	builds merc.
6	2110BC	builds worker	builds worker
13	1830BC	-	builds merc.
17	1685BC	builds granary	-
19	1635BC	-	builds settler
23	1535BC	builds merc.	-
26	1460BC	-	builds merc.
27	1435BC	builds settler	-
32	1310BC	builds merc.	builds settler
36	1210BC	builds settler	-
40	1110BC	-	builds merc.
41	1085BC	builds merc.	-
45	985BC	builds settler	-
46	960BC	-	builds settler
50	860BC	buidls merc.	-
54	760BC	builds settler	builds merc.
Total merc.	=	5	5
Total settlers	=	4	3

1. As you can see, after building the granary, Legopolis produces a mercenary+settler pair in about 9 turns, while without it in 13-14 turns.

2. The time needed to build the granary will be practically recovered in ~ 40 turns which seems to be a lot, but ...(see below).

3. What the table doesn't show is the fact that Legopolis, with the help of the granary, will not only have a bigger shield output (growing faster and thus being constantly bigger), but also a bigger gold and science output.

IMHO the granary is definitely worth it.

[lmtoops]

Wow a very compeling argument back with solid facts. You must not be a politicain.

I'm all for the Granary.

[vondrack]

Good work, Tibi! 40 turns is a reasonable time, considering how long it will take to settle all of our landmass. We will definitely benefit from a granary in Legopolis greatly in the long run.