Cradle 1.2 farmer test

[Hermann the Lombard]

Hexagonian mentioned that he isn't sure just how effective farmers are, so I ran a little test (and I mean little). This may prompt more serious experimentation. I took my starting position (Assur's ending position) from Succession Game V. So I'm starting on the 81st turn. The conditions are as follows: Tyranny, Rations:4, Workday:12, Wages:8, PW:20%, Sci:50%, no buildings, no tile improvements. I then ran 20 turns from that position, twice, once using no farmers and once using maximum farmers. The numbers are in the following order: Pop/Food/Prod/Gold/Sci. "After" means after using maximum farmers, but with all citizens set back to workers for comparison with "Without" (and "Start" is Turn 81). Tanis founded 5280 (Turn 87). Ending position Turn 100.

Heliopolis

After: 5/76/71/10/12
Without: 3/54/52/8/11
Start: 2/42/43/8/10

Tanis

After: 3/52/43/10/14
Without: 2/43/35/9/13

Combined
Change : 3/31/27/3/2

So with the farmers, I ended up with 8 citizens vs 5 without, which is significant. But under tyranny, the yield from the extra 3 is minimal indeed: 27 production, 3 gold, 2 science. In the meantime I produced one less warrior in Heliopolis and didn't produce even 1 spearman in Tanis (no rushbuying). Would one less warrior be fatal? In Cradle, it might be.

I also looked at Memphis and the recovery after producing the nomad that founded Tanis. The population drops from 6 to 5. Should I let the population recover "normally" or should I use farmers to rush the process? Here are the numbers with 0-5 farmers, respectively, with the number of turns needed to recover the citizen:

0 farm: 5/99/61/15/20/17 turns
1 farm: 5/113/52/13/16/13 turns
2 farm: 5/126/45/9/12/11 turns
3 farm: 5/140/36/6/8/ 9 turns
4 farm: 5/153/28/3/4/ 8 turns
5 farm: 5/167/19/0/0/ 7 turns

0 farm: 6/114/69/17/22
Gain: 1/15/8/2/2

So, you can cut the time required by 60%, but during those 7 turns you are getting zero gold and zero science from this city, and you gain only 2 gold and 2 science per turn for your trouble...and only 8 production, after giving up 42 production for 7 turns.

I'm still inclined to spit out 2 or 3 defenders/explorers from each new city, then use farmers to bring the population up to 5 or 6, but I'm giving up a lot of "yield" along the way. Now I'm presenting the extreme cases here, and you could get intermediate results by using fewer farmers than possible. As always, there are tradeoffs.

-- Hermann

[Locutus]

Interesting tests. I basically agree, I think very early in the game you really need production ASAP to pump out units for recon and beating off the barbs, so little to no farmers ought to be used. For somewhat later on in the game, it would probably be the best strategy to go for something in between, 1 or 2 farmers or so (while the size is 3-6, anyway)...

This kind of number-crunching is really interesting to do, we ought to do it more often. I very much enjoy it myself but unfortunately rarely have the time

[stankarp]

I found in CTP2 Gameleague games no one used farmers. They really slow the science and science is critical. I tested it a few times and decided NEVER to use farmers, mainly because of what they do to science. In my current mammoth game, 15 civs, extra hard on a gigantic map, I only use farmers occassionally to balance labourers in frontier cities churning out combat units.