Opening moves

[akots]

It seems that opening moves are extremely important in Civ4. There are a few crticial step that determine the outcome and these steps are done very early in the game.

There was some previous discussion on when to build the first worker and settler and how to ensure growth on higher difficulty levels starting from Emperor. It would be nice if we can somehow organize this knowledge. It is a painful process however.

I've been experimenting a little bit with early moves on Emperor as Gandhi on a duel map having a continent of my own against a single AI (Genghis Khan). No barbarians and moderate climate.

Depending on the presence of local non-calendar luxuries and other goodies the start is either playable or unplayable. For starters, there is a limit of 4 happiness for growth which means that without additional happiness, the capital can grown only to size 4. Health is also important. However, if the city becomes unhealty, it does not stop growing, just one food is subtracted for each unhealthy icon exceeding healthy icon.

There are a few questions:

1) When to build first worker? Consider that the start is with cow (or wheat) on a grassland. Otherwise, it is pretty much unplayable imho. If start is near wheat, Agriculture has to be researched. If start is near cow, agriculture is still useful but then Animal Husbandry has to be researched. I then researched Meditation and converted into Buddism which gives plus one happiness. Usually at this point, wheel has to be researched so that health or happiness resources can be connected. I have started to build worker first at size 3 and this allowed to grow to size 5 before building the first settler. Meanwhile, I've built mostly warriors and sometimes temple.

2) When to build the first settler? I started to build the first on at size 5 with production time around 10 turns for the settler. Then, again depending on the surroundings, either continued to grow the capital in size if happiness (health) permits or kept producing settlers. This gives a 5-city empire around 1000BC with some limited military. Byt this time I can usually get to Monarchy through Priesthood (popped it from a hut twice in two different games; yes, it is possible on Emperor). Then all cities can grow to size 5-6 or even 7 sometimes due to state religion spreading and military police. Still AI somewhat outgrows me albeit very slightly.

Now, I would be extremely curious if building settler before first worker is better? Is granary useful on that level of difficulty? What other research paths are more reasonable? Is there any way to catch up with the AI techwise or it is hopeless? Are early wonders worth a shot?

What is a best civ for early growth? India is good due to fast worker but mining is rather useless early in the game. Mysticism is extremely useful though since it is possible to found Buddism early. I was thinking about Inca who start with agriculture but their leader is kinda useless (aggressive trait).

[ProPain]

Like I said in the Q&A thread, it should be possible to make some calculations based on on some assumptions about bonus resources. It would be very interesting to see what is the quickes path to say 5 cities. I'll try and look into that this week, today won't be possible I have to work untill 21:00.

Calculations also should say something about the worker/settler first question. My gut feeling says settler before worker is better in general as :
- research is needed to build the necessary tile improvements.
- 2nd city will add another production queue, having that available sooner should be benifical.(iirc Usually at size 3 worker takes me 15 turns and settler 25. So building a settler 1st means approx. 10 more turns on your 2nd production queue I guess.

But again, this is gut feeling some serious mathmagics should shed some light on this.

[grs]

Since most improvements on resources are extremely powerfull I tend towards a worker first. I have no hard facts though.

[Pastorius]

usefulness of worker depending on starting land and starting tech to some extent, also on tech path: certain starting techs allow some worker operations. Resources may vary, and tech path may vary - though I suppose the religious path (meditation or polytheism) is not always chosen as first step.

therefore I guess PPs generalization is good, considering that those tiles and techs vary more than the usefulness of a second production queue.

and the relative proximity of neigbours strengthen this - all other things held constant, a settler seems more appropriate - even though landgrabbing is more expensive in terms of maintenance cost, waiting too long means they box you in, also taking into considerations that most empires are smaller than in civ3.

[akots]

quote:Originally posted by grs

Since most improvements on resources are extremely powerfull I tend towards a worker first. I have no hard facts though.

It seems this is a good idea although not substantiated with numbers. Every time I made settler first on these duel maps of mine, I've ended up greatly behind in grown. However producing worker at size 3 made the trick. I even decided to continue the game through the night (played till 4 am) against the evil Khan.

Well, at 10AD I've had 7 cities versus his 6 and have collected most of wonders on the Classical Age and some of the Reneissance Age. The bastard however went through the military route completely neglecting Drama, Music, etc. A particular thing about that duel maps is that AI will never trade to you anything, no techs and no resources. He was however trying to extort something constantly. The peaceful coexistence ended when Khan landed settler/worker/longbowman on some of the tundra tiles and built a city there. City had been taken through military means and this did not improve our relationship...

Since I had no iron (only was able to build about a dozen elephants and same number of warriors and cats), me and the Khan played to Gunpowder/Chemistry. He knows Rifling and has rifles all over the place with about 3 promotions (wondering how did he get them without barbarians) and I have grenadiers+cavalry+cats trying to beseige his cities. Well, if only I had cannons, that would be an easy task. But cats are really weak against riflemen. So, we'll see how it goes but I firmly believe the Khan is doomed. It is just a question of time and patience.

I believe I had firmly outgrown AI during the Ancient Age and even managed to produced about the same number of great people. However, I have somewhat overexpanded by building 2 rather remote cities trying to control calendar luxuries and then the growth stopped because of budget problems. It then continued and now (1810AD) I have 10 cities against Khan's 6 or 7, not sure about that. Since this is duel size map, maintenance should be extremely high and it was possible to grow rather large still.

After this sudden increase in maintenance, I started to closely monitor the state of each city. If the city was due to grow but did not have enough happiness or health, I have either built a military police for happiness or tried to improve health with granaries and aqueducts. Having Hanging Gardens also substantially helped. It is a lot of micromanagement and sometimes it does not work like you think but still I'm having most of the cities grown to maximal size they can afford and the capital us at 19 health and unlimited happiness (due to Theater Wonder). Other cities are around 16-17 health and 12-18 happiness depending on improvements.

[Lt. Killer M]

quote:wondering how did he get them without barbarians)

seen odd promotuions, too: he is NOT in a civic to get free exp points, lacks a res. hooks it up - voila, two promo new units exist one turn later.


upgrade??? not when you are flat broke....

[Pastorius]

Akots, when happiness is a problem, turn on avoid growth till you solve the problem?
Not perfect, but less micromanagement (accessable through "mini-city view")

Seems to be under testing in some SGs @ that other forum. (LKender among others IIRC)

"Correct" civic and barracks count for 2 promotions iirc. Where did he get the last one?

[akots]

I think the last one he got from his trait which is aggressive.

The whole point of the game is to let your cities grow non-stop until they reach the limit imposed by food constraints. Now, instead of playing around with the governor settings, it might be a better idea to go and find some health or happiness out there to sustain growth. But this is LK style and there is no point to argue with that since the man is solid like a rock, all arguments bounce off. Never mind, LK is a VERY nice guy, he's just got somewhat peculiar playing style.

[Beam]

Couple more arguments to let cities grow first, then build settlers/workers:
- Maintenance increases almost exponentially with the number of cities, i.e. in 4 size matters more than numbers with respect to cities. Courthouses (120 ) are needed pretty soon to keep cost in control besides measurements to generate gold like a religion.
- There are no hard limits to city size such as aquaducts and hospitals like in 3.

Also, tiles in the inner 3x3 can't be used by other cities, so it does not make sense anymore to build the core in a city-tile-tile-city pattern. You'll need to build wider anyway to get enough landcoverage and access to resources. City-3 tiles-city is a much better pattern and in case of poor land city-4 tiles-city as well.

[Kemal]

Congrats to akots for doing so well on emperor, not a small feat.

However, risking to sound elitist here (what I do not mean to), I do think the difference in dificulty and approach between a 1x1 map and a "standard" game is so big that comparing and utilizing the same opening moves is nearly useless.

A standard pangaea map with 7 AI opponents seems to take an entirely different approach. Getting a religion founded without a spiritual civ is almost impossible, so you can skip the happiness of a religion. Also, it seems that on maps as these, you will usually find yourself without either horses or a metal resource, and with only 2 to 3 lux resources max. Furthermore, beelining to monarchy means certain death it seems, as you are guaranteed to be engaged in a few wars with your opponents before 1000 BC, and we're talking 10+ units per attacking civ at least.
And I'd like to second akots' comment on difficult trading, this goes for multiple opponents as well... you really need to invest big time to get some trades going. Unfortunately, the AI makes friends a lot easier, and they seem to manage trades a lot more efficiently. Might be that is my ineptitude though, of course.

I've been playing quite a few of these games, getting mopped up quite a bit. One thing that I'm now sure is required fast in games like these is archery, and at least 3 archers. These archers will need to fight the brutal barbarian onslaught and get at least 2 promotions to serve as backbone against the AI hordes, until you can secure either a metal resource for spearmen/axeman, or manage to reach HBR and get Horse archers.

Now, I am only playing the beta version, so I'm not sure how much of this is applicable to the retail version, but to illustrate how much of a problem the barbs are, here are some combat stats from 1500 BC, when I had managed to build 3 (almost 4) cities:


6.88KB


3.57KB


3.58KB

Furthermore, I was attacked by 5 chariots and 2 warriors next turn by Montezuma, which as said happens to me every time on emperor.

Looks like heavy investment in military is sure worth it. Now I have to say this was with Persia, without horses, so no horsebased unit or immortals for me, but even then, I think archers are a must, since an archer usually beats a chariot with much ease in defense. How to prevent them from pillaging my resources all the time is something I still have to find an answer to though.

Finally, on the worker/settler issue.. definitely a worker for me, as grs pointed out. Second city can only grow to size 3 anyway, and without a worker the land it will use will be poor anyway... also, repeating myself, I need a powerful city to churn out a few defenders before daring to venture into the fog of war with a settler. Landing next to a bear or lion = or > 50% death with just a warrior as escort.