More Domination Theory

[BlackBetsy]

Well, after finishing off a very satisfactory GOTM41 (my highest score and best performance by a mile out of COTM9, GOTM40, COTM10 and GOTM41), I went back to Dominations.

I'm fascinated by the Tiny Dominations for their impact on speed. I've submitted two more, one a #1 spot, the other a good finish, but too slow to be among the top spots in Tiny Chieftain.

It strikes me that, when it gets down to it, there are about 3 conditions for a really fast Domination:

(1) Settler popped from an early goody hut;
(2) Early SGL (for ToA); and
(3) Starting location that is a 4 turn settler factory.

There are three Civs I can think of for a good early start:

(1) The Celts

With the Celts, finding a 4 turn settler factory is easy - 1 FP Wheat is all you need with the city square bonus. 3 bonus grasslands and a forest, and you're golden. Taking the Celts means that you have a lower chance at an SGL, though.

(2) The Persians

The Industrious characteristic means better workers setting up roads and settler factories faster. The Scientific trait increases chances at SGL. The Bronze Working start means you get the leg up on Iron Working for the Immortal UU.

(3) The Babylonians

The Scientific trait is important for the SGL concept, and the Religious trait shortens the inevitable revolution to Monarchy. Also, the early offensive UU (the Bowman) is a more controlled way to trigger a GA. You might need a sword or two to fight an early battle, and having the Swordsman be a GS or an IMM means you might prematurely trigger a GA.

Even more than all three of these, however, is a good map. Last night, I played with the Babylonians on Tiny Chieftain domination. Flood plain wheat and other flood plains in starting location. Check. First goody hut, 2 squares from capital was a settler. Check. Mysticism produced an early SGL. Check. ToA built very early, and had two settler factories (4 turn and 5 turn) cranking out settlers very, very early, and had 7 swordsman ready early.

Here's where I got bogged down - the map. I actually had to drain a couple of marshes (I had the friggin map set to arid!) and the Pang was lonnng....and had two major islands to thwart the Dom limit! My victory was delayed by having to get settlers through a narrow isthmus into the bigger half of the continent, and having to settle one of the two islands. ARGH. Domination in 450 bc.

I was stunned by Smirk's 1450 BC Domination victory on Tiny Chieftain. I loaded it, and then saw - it's the most beautiful Tiny Pang I've ever seen. Big enough that he could allow 4 AI cities to remain at the end of the game and still hit the Dom limit!! He used the Persians, so speed in roading was not an issue. Other civs were Iroquois and Mayans....which struck me as smart because their cities were MUCH more likely to rise to size 2 and not be auto-razed. In a fast domination, you WANT the other civs to have a lot of cities - it saves settlers, and allows you to grab territory faster.

I'm thinking that the fastest domination might occur, however, on a perfect (like Smirk's) Tiny Monarch, Emperor, or Deity Pang map. The AI needs to build as many size 2 cities as possible, but not be too strong to defeat the 7-10 Gallic Swords or Immortals it takes to complete the domination. On Chieftain, the AI's growth is so darn limited. I can see such a domination coming before 1450 bc, but not much more. Smirk may have thrown a perfect game there on Chieftain. There probably is a perfect game out there on Monarch or Emperor that gets you to 1600-1700 bc...but I can't see anything earlier.

I will try to use ag civs as AIs for their early growth. Hard not to use the Celts for the fast Gallic Swordsman, but the Industrious bonus of the Persians and the higher chance at SGL may seal the deal.

 

[boogaboo]

What about a SGL for early Pyramids?
This seems more important at the start then ToA, which can wait a few advances...

 

[killercane]

Well Im pretty sure that 1600 BC is not doable. You need the pyramids and ToA via SGL, multiple settler factories, a small domination limit, a strong UU to beat the Monarch and Emperor spears, at least 17 cities if you look at Smirk's great game (I think he had 21 total?), no sneak attacks by the AI on your undefended cities, and manufacture enough military to take out 4 AI cities in one fell swoop 5 turns before you plan on winning. Thats a lot to ask for in a game.

On Mon and Emp the AI aggression level is higher and they build more and better units quicker. I think a useful technique might be to use a Boogabooian Jag warrior rush to take out one AI by 2500 BC, get a few cities, and keep expanding all around while waiting for the other AI to get 3 size two cities. But you lose probably another city that the vanquished AI would have built, and using your jags against civ2 will be tough against spears.

A rundown of UUs/civs gives a short list of possible candidates for dominating by 1000 bc on Monarch:

Jags- agr and a two move UU, but tough going against spears.

Egypt- industrious, two move 2 attack unit one tech in, start w/ CB so closer to ToA, but not Agr.

Iroquois- agr with a two move 3 attack unit, but the unit comes 3 techs into the timeline + you need to go 3 techs in to get polytheism.

Sumerians- agr + sci for better chance @ a leader, but a one move one attack UU.

HIttites- UU too costly among other things like unfavorable traits.

Persia- a stud UU available one tech in, industrious, scientific, but not agr for the quick granary.

Inca- UU costs more than a jag and expansionist is not really needed.

Celts- great UU and rel/agr, but high cost to warrior upgrade em and you cant hold back settler production in your high production cities (usually your settler factories) to make new ones. They lose vs. spears sometimes as well.

You'll never change gov'ts unless you are religious and even then only monarchy in all probability. I like to get to writing if I dont have an SGL after researching polytheism, then you can get philo + something else which doubles your ability to get an SGL by discovering two new techs. I dont think you have time to research a 30-40 some odd turn republic.
I think mon is doable by 1200 and emp by 1000 w/ persia, Montezuma's warriors or Egypt w/ great maps.

 

[killercane]

I forgot to mention that if you declare while playing w/ persia or another one move UU the AI will whip a spear in their size two city before you can attack, which is not good for obvious reasons. Every little city is huge for a good end date.

 

[BlackBetsy]

I forgot to mention that if you declare while playing w/ persia or another one move UU the AI will whip a spear in their size two city before you can attack, which is not good for obvious reasons. Every little city is huge for a good end date.

Yup. This is why the Celts are pretty fabulous for it. If there was a scientific two-move UU civ (can't think of any), that would be the one.

Can you pop two settlers from huts? Popping two settlers and getting a GSL might be the ticket to a pre-1500 dom. Awfully hard conditions.

Continues to make me think that Smirk threw a perfect game.

Just played against the Mayans. Ugh, they are tough with their UU. Maybe the Dutch (Agr, seafaring) would be another good choice with the Iroquois.

 

[SirPleb]

I forgot to mention that if you declare while playing w/ persia or another one move UU the AI will whip a spear in their size two city before you can attack, which is not good for obvious reasons.

They can even do this when you have a two move unit attacking - they seem to get a chance to whip in the middle of your turn at the point when you declare war.

There is however a tactic to get around this: just before attacking them gift them Monarchy. They should be in anarchy when you attack and unable to whip

 

[killercane]

@ BB-
"Can you pop two settlers from huts? Popping two settlers and getting a GSL might be the ticket to a pre-1500 dom. Awfully hard conditions."

Only if the AIs are outexpanding you, the chance for a settler is based on total number of cities including the AIs. You might get one at the start at 3950 and maybe one later but unlikely.

@ Sir Pleb-
"They can even do this when you have a two move unit attacking - they seem to get a chance to whip in the middle of your turn at the point when you declare war.
There is however a tactic to get around this: just before attacking them gift them Monarchy. They should be in anarchy when you attack and unable to whip."

Absolutely, a good point. Thats why Im thinking that doing the research from the wheel to the philosophy slingshot on Emperor and Monarch, when playing as the Iro, would be prudent. Playing against two religious AI w/o alphabet, Im going wheel, writing, code of laws, philo-republic. If no SGL yet, research is set to min on HBR after a warrior code trade, produce chariots, upgrade then attack. The AI beeline to polytheism so I can trade for that if I get an SGL; otherwise I have to wait for them to build it, which does really slow progress.

What do you think about revolting to republic? I dont like to waste five turns, even for the increased benefit since at this time you should be wiping out AI cities.

 

[DaveMcW]

I don't think Republic can pay off for 1000BC domination. In addition to the production loss, the reduced worker speed during anarchy hits just when you are building the roads for your final settler push.

 

[BlackBetsy]

What I haven't tried is getting writing fast and then doing a double-ROP kill on the two civs. Although ROP is kind of problematic in the early stages with settler/escorts wandering through - the escorts could grab towns behind your lines.

Gifting them Monarchy the turn before seems to be a great gambit to prevent the AI from pop rushing an extra spear.

Regardless, getting them in Monarchy will (1) disable pop-rushing (they'll have to buy units); and (2) helps their food production - more size 2 cities.

Seems to me that a Monarchy gambit makes a lot of sense.

Of course, someone is going to probably tell me that the AI can pop rush in Monarchy, too.

 

[Smirk]

I did a write up of my game and its down the page a little, here is the link which may work: http://forums.civfanatics.com/showthread.php?t=112495 .

I was stuck in a republic path for some reason, but yes researching either has only one purpose in this game type: gifting to the AI to stop them destroying their cities (and as an aside less defenders). I mistakenly switched to republic in some other attempts and its a waste, its positive long term effects will not appear in the time period these games are using.

I think I could improve on my date and my writeup discussed some ideas along those lines. Unlike many of the other games of this type you have to do some early mongering, like the post says I had 5 cities after I founded my first settler built city, and this is not at all slowed compared to a non-mongering game. Some of the other top games did capture another city, but 5 cities after 20 turns is alot stronger than 3.

Also, the celts can potentially give a good game but the UU speed is not useful in the sense of earlier or quicker unit deployment, its useful in the same way the immortals higher attack is useful, you need to build less to get the same results (since they retreat). And thats the important thing you want to be able to build enough but not have to spend lots of valuable early production on them, and really you do very little warring in chieftain, its only a few extra cities. I think the agricultural trait easily outweights the higher production cost of their UU, and the lack of the scientific trait can be overcome if you play enough games. So the only problem is you need a few more workers to build roads. I don't think thats a big deal since industrious is pretty weak compared to PTW, you would need double the workers in PTW to compare, whereas in C3C you only need 25% more or something.



I have some other more concrete ideas how to improve this game type, but I'll wait till I get a chance to play them, I don't want a repeat of this month where my plan is used and improved upon before I even get a #1 spot!

 

[BlackBetsy]

Wow, Smirk, great write up.

The thing is, you've got a GREAT Pang on that map, with the 292 dom limit. I've never seen a compact Pang like that in the 50+ games I've played long enough to see the shape of the Pang(sometimes, even after a promising start, I recognize the Pang is too ugly and quit).

I could see spending the early SGL on the Pyramids, too. I've got an SGL saved in a game I'm playing, and I'll see what the difference is between ToA and Pyramids.

Two SGL's is almost too much to ask. Took me my last 10 games or so to get one.

Writing/Philosophy seems to get me more SGL's than other techs. Weird.