SGOTM 13 - Plastic Ducks

No, at t71 you get to size5 and work the GH mine, -1F+3H.
I forgot to change the other to 9f surplus, thanks!

The no-whip approach vs T62 LH whip

This one is interesting

Legend: (only accounting capital)
turn - food in the bin - food surplus - city size - total hammers produced - total commerce produced - hammers/turn - commerce/turn


The difference is (no whip vs whip)
+17F-24H+2C

17 food is significant: that's 2 turns earlier growth based on a 9F/turn surplus.
It also lets us 2/3-pop whip the marble settler faster should we wish to do so.
OTOH, without a granary, the food conversion to hammers isn't that great.

If we had a granary the choice would be obvious but we don't...
24H/17F is 1.41:1 conversion rate, basically a whip from size 2 to 1.

I'll run a few t107 tests with both approaches - it's probably easier than trying to correctly interpret these numbers. I haven't tried too much the MC approach so far but I think Snaaty is doing great on that end.

After further consideration, do we really want to research through Polytheism? If anything, that is one of the techs I would grab from an AI with a trade. Plus if we were to grab the religion, there'd be less chance for religious strife between the AIs...

Do you mean skip Monotheism? Or do you mean go Med->PH instead of Poly->PH?

Got it.
I didn't notice we could complete the galley on T61.

Which makes this post irrelevant.

Compromise said:

Do you mean skip Monotheism? Or do you mean go Med->PH instead of Poly->PH?

Click to expand...

Skip Mono and Poly.

Even if we end up taking MC as freebie, the cost of Forges is exorbitant:

180H/.25 = 720H to pay back = 6 Maces, +1.5 for the cost of the forge. 7.5 Maces per city is a very large number! {Maces are 120H? I forget}
180H/1.25/.25 = 576H+60H (missionary) = 636H to pay back. = 5.3 maces +1.2 for the forge = 6.5 Maces
I don't consider buildings in this situation because, we fail hard if we do go for infrastructure.

We'll see if forges are actually useful I guess. It depends on how much time difference it makes on our warring dates (earlier warring = better)

I'm wondering if all-out-conquest is not possibly faster than Domination... if we strike early enough, AIs will not have many cities and we can win faster than by acquiring the necessary land for domination???

Maces are 105H
Xbows are 90H

If Astronomy is not needed, we may be able to be fast. Can we tell if the AIs are all near each other on a Europe-like continent or if they are separated on their own islands?

For super-early, let's Oracle->Construction and start whipping catapults from all the seafood around here....

Ok great, thanks. That makes it even less worth Forges in most cities.

We can't tell right now if the AIs are clumped up together. I'll try to do some more camera flying when I get home but it's really looking in the dark with regards to what the AI is doing.

@ kossin:

more micro details for our capital:

even better then whipping the lh in 62 is whipping the galley right now (61) and finish the lh with the ov. not doing so will result by 1 for 1t in our capital with max growth... ...i might have mentioned before, but im sort of a micro addict in demo/teamgames... ...sometimes a single hammer does wonders

...

sorry i havent mentioned the above before you did your tests kossin...


oh, and in the end i opted for not whipping the lh at all and whip the settler pop3, which even was better

...

working on the full testmap right now btw...

...

forges:
we basically need 2 forges if we take the mc pick (still assuming we have copper):
1. for breeding the ge. free machinery is too god to be ignored i guess
2. for the colossus

if we go for mc, i would strongly recommend taking mono. research would be:
myst-masonry-pottery-poly-pries-mono

building i would the first forge in the holy city (org bonus) and the second one i a city that the reli spreads to. if not, the colossus should go into the ge breeding city, poisoning the pool a bit. but if we d.)raw a gm, no big deal, we run a trade mission and will be fine (or research cs and use him to bulb guilds...)

forge whith mono is: 180*0,75=135 hammers which equals 105*0,25=26,25
135/26,25=5.1 maces, lets say 5

if the build in each city is:
lh-gran-(forge)-baracks-(stable, depending if knights are worth it) then its even a bit better. and i assume we will whip at least 5 units per city and we als need ships, to transport units, worboats, settlers and so on, so i will pay back in time

You mean whip the galley next turn?

It delays the gold city 1 turn and we can queue a settler on top of the build list in Paris the turn before growth to avoid the unhappy.

+3F-3H-2C in capital
-5F-1H-2C in gold city
-4C from trade routes
-7C from 1 turn later gold mine

What am I missing ???

BTW, I love microers I am one myself, when I have time!

EDIT: Actually, it`s only +1 food in the capital since we get to work a clam 1 fewer turn (3-2=1)

oh, and optimal city growth we will have going down that path:

myst-masonry-pottery.

like that we can build in all cities lhs then grans, including a gran in the capital after the settler for marble city. it wont even delay the next settler or the oracle, cause of the faster regrowth rate. we gain quite some hammers by whipping early with grans...

...

edit:
forget what i said about whipping the galley. its wrong info based on one of my older tests

dont whip the lh, just finish it normally and do a 3 pop whip on the next settler. should be best (at least was in my approach going myst-masonry-pottery)

when will the stupid uploading thing be fixed btw? does anybody know?
i think i have worked out a close to ideal play for picking mc t101 and would like to share the save at some point... ...but i guess at a certain point i will have to list all cities with all builds and such anyways... ...urgh

full testmap is also ready. have fun killing the european ais

havent played up to turn 60 yet in the full testgame. everybody feel free to do so and post it here (if possible).

nappy: small testmap based on kossins map
nappy full: same starting area as in kossins map but with "europe" near and waiting to be conquered

maps emailed to:
compromise
kossin

For certain, a Machinery bulb is interesting with a GE. I will look into the various bulbing gambits we could do.

kossin said:

For certain, a Machinery bulb is interesting with a GE. I will look into the various bulbing gambits we could do.

Click to expand...

i have downloaded a bulbing sheet ages ago, if anybody is interested (you know the trick now... ..pm and you will recive)... ...from a quick first glance, mainly the ge-machinery or a gm-cs/guilds might be of interest. it might also be interesting to put the colossus in the ge breeding city, making it someting like 80% ge 20% gm...

...

i have quite some spare time this week and maybe even the next week. so anything that needs to be done, just feel free to ask...

It's [reference sheets for bulbs, etc] also available in my signature under the 'AI behavior' link.

If we were China/Native America, there'd be little doubt that GE->Machinery would be great. It can still be good here to give us some pretty early Maces though... food for thought.

some flaws in full-testmap corrected. played up to turn 60. please use the new save...

I have an Oracle -> Currency on T106 (through Polytheism).
My computer just rebooted without warning, so I won't post the exact micro just right now. But maybe we're not interested in that anymore ?

Basically :
- Paris : no Lighthouse whip in Paris (not T62, nor T63, slow build), which allows to build a settler at size 5 (no whip, health cap : max output). Then Paris builds the workboat for Lyons, then stores a Galley & Settler for 2pop whips with max overflow.
Paris grows to size 7 on T89 without suffering unhappiness, which allows to work some some coasts.
Paris' growth : grows on mines until the workboat for Lyons is complete (size 6).
- Orleans : goes LH -> WB -> Warrior -> Settler. It works the gold each turn (governor tends to work food when doing binary research if we don't assign the gold manually). The warrior completes with the growth at size 5, when the settler is started.
- Lyons : starts a WB when it is settled (will improve the fish). Lyons is settled on T82 and the clams are improved on T88.


Thank you Kossin for your post n°141 (Lighthouse whip analysis). I really got confused with the turn of completion of the galley (available right now! not in 2 turns). And your post opened my eyes.
Having the gold available 1 turn earlier makes a huge difference.


Hmmm... That said,

Oracle picking Currency :
Mathematics has a serious opportunity cost in terms of scouting.

Machinery bulb :
I'm very interested in this. I've never done such a bulb in solo play and it feels more in the spirit of this game. I haven't tested the Oracle -> MC path, yet, but I will do so shortly.

Marble wonders :
I'm not sure they are so useful. Except for the 2 Heroics. The Parthenon is appealing to get Golden ages but the Great Library... is it worth the delay ? I don't know... However I think that the 2 Heroics alone justify our trouble going through Polytheism rather than Meditation on our way to Priesthood.
There is also Monotheism that is opened by Poly.
Finally, Meditation forces to bulb Philosophy before Astronomy. If we need Astro (in which case, the Great Library's value rises), then we're also better off having researched Polytheism.

Conquest vs Domination :
I have no experience of Conquest but, Vassals being turned off, I guess it is indeed faster. Reaching the needed land % could be troublesome for Domination victory.


@ Snaaty :
Great news about the test map ! Thanks a lot.
I'm sending you my e-mail.

Thanks for the save, Snaaty.

Slight forest placement error, easy to correct:

Copied this pic from srad's post:



The three-tile island directly west of Capital Island has no forest in the middle (test game has an extra forest). The tundra island has a forest in the west tile of the southernmost row (missing in test game).

I am...not fast at playing, so will see how far I can get.

@BornInCantaloup: If Snaaty's gone to bed, and you're itching for early test save delivery, PM me and I'll send his files to you.

I had that same exact forest grow when I prepared my save, might be the explanation.

@BIC
Nice test, if you ever manage to put the micro down, I'd be interested in seeing it!

both testmaps revised again (now proudly presenting v.1.2) and played up to turn 60:

nappy 1.2: small testmap based on kossins map, no ai contact
nappy full 1.2: same starting area as in kossins map but with "europe" near and waiting to be conquered

maps emailed to:
compromise
kossin
bic

From my notes :
- Research : binary all the way until the last 3-5 turns. We have some margin to reach Mathematics in time and different options yield equivalent results. Not sure what the best partial slider is.
- We need to be careful with Orleans, manually assigning the gold tile. Otherwise, the city will switch to food when the slider is 100% gold. Its build orders are LH->WB->Warrior. The Warrior is completed the turn Orleans grows to size 5 (no trick). Then it stats on a settler.
- Paris slowbuilds LH (corn > clam > mine > coast) and reaches size 5 (no special micro). It then starts on a settler.
- Worker use : improve the gold, then chop the forest in marble island, then roads+quarry the marble, then heads to the fish hill.
Then :
T76 : Paris works 1 coast + 1 mine (max overflow into next building).
T79 : Settler out, Paris starts on Workboat. Works 2 mines.
T82 : Lyons is settled. Starts Workboat.
T83 : Polytheism -> Priesthood.
T84 : Paris is size 6. 1 Unhealth.
Workboat is complete (-> clams). Starts on Galley.
T87 : Priesthood -> Writing.
Paris starts on the Oracle. Works 3x coast.
T88 : Paris resumes Galley. Works 3x coast.
Clams are online in Lyons.
T89 : Paris is size 7. Starts settler working 3 specials + 4 coasts
T91 : Paris works 2 mines + 2 coasts (I'm not sure about the exact micro, here. Will be checked. The idea is to whip the settler on T96 to have Rheims settled on T99).
T94 : Writing -> Maths.
Lyons is size 2.
T96 : Paris whips the settler.
T97 : Paris resumes the Oracle.
T99 : Rheims settled. Works fish.

From then on, I don't have detailed notes. Here is how it goes, though :
- A workboat from Lyons will improve the fish.
- Paris keeps building the Oracle, maximizing food and commerce (coast > mine)
- 2 turns from Maths, Paris resumes the galley (ready for max overflow) and whips it into the Oracle.
- T106. Currency is in. We have 0-10 gold in treasury (I had 8 in my first attempt, meaning we can afford working less coasts when we're not building up a whip overflow).
Paris is size 5 and has been stacking anger very recently -> there is one unhappy. The Marble is online on T90 and our first whip is T89, so I'm not sure whipping earlier is better.


Some details may be off. Especially regarding the settler whip between T89 and T96.
It seems whipping on T96 doesn't give the best overflow and I can't recall whether I microed or not. Microing should delay the settling of Rheims to T100 but... well, I forgot.
Microing for max overflow is : 2 turns working 4 coasts + 8 turns working 2 mines + 2 coasts + whip. 10 turns are needed before whipping.
3pop whipping the settler is a possibility I haven't tested yet. We could do so with 5 turns of production into the settler (T95).
I'm not convinced it changes much to whip anger, though. Growing the capital to size 7 is (I guess), what allowed to shave 1 turn for the Oracle.

Production is very poor, though. On T106, we have 2 galleys out so we can expand, but we haven't had much time to do so.
I will try to clear this sometime today