Rush pitfalls

[Mantic0re]

This is probably my biggest self-assessed weakness because I absolutely abhor warfare.


So in my last few games it quickly becomes apparent that I'm in a position to be boxed in with less land than I want. I can squeeze in 9-12 cities but the truth is only about 6-7 of them are on juicy sites without tons of desert/tundra tiles or so food poor as to be negligible pre-Biology.


In all of these games it seems obvious at some point that a rush is called for. Some of these games I've replayed in an attempt to improve my position or victory date. I have to get over the hurdle somehow.

I've tried Axemen and HA rushes against a few AI's. I've read that Axe-rushes are best done with 2 cities and 3 workers (early). I've not seen anything written on HA but I tried them with 5 or 6 cities and 9 workers. I'm not terribly happy with the results. I know the efficiency of build-up is a key limiting factor and improving this makes for better outcomes. Still there are some key things that I think absolutely kill the rush idea outright (at least at my skill level).


I tried a rush across something like 22-25 tiles of mostly jungle. Bad idea. (I remember being boxed in and where the other civ was) Pre-roading helped but the hammer investment was surprisingly steep. Turns out it was better for me to take my narrow strip of land and tech until catapults than to rush with axes over such a long distance. I didn't have horses or I'd try that. So I'm thinking there must be a maximum distance that early wars can be successful in.


The hammers invested in rushing Toku doesn't seem worth it either. I thought if I chopped more I could make up for the bonuses but it didn't play out like I hoped.


These attempts were the most troublesome. The other attempts were more obvious (too me at least). HA vs. Egypt worked well, avoid axes. Immortals and Praets are to be avoided. Cutting the resource still costs a few units / UU to down and then you've got a pissed of neighbor designed for early warfare.


So for now I throw out any idea of early warfare in the following conditions.
1. AI Capital more than 15 moves away before roads.
2. UU that counters primary intended attacker.
3. Protective AI.


If anyone has additional criteria they use to take pre-catapult warfare off the table I'd love to hear it. Please feel free to add caveats or critiques. I'm always open to advice.

 

[TheMeInTeam]

Rushing super anti-rush AIs that have protective + high unit spam or AGG + high unit spam isn't a good idea, and neither is attacking AIs 15+ tiles from your capitol!

Just wait for a more opportune time to attack. With 6+ cities (standard map) it is very easy to bide time until renaissance or later and capture cities with less relative investment than murdering your tech and risking the game to rush across vast distances. If you have 8-12 cities you can often viably win on most difficulties on standard world sizes without war.

In short, when they're that far away protect yourself but spend hammers more economically than trying to rush at extreme distances.

 

[Mantic0re]

I've noticed after replaying a few different positions that if I build units in preparation for a war then I find the AI builds more units and if I'm going for a generally peaceful strategy they tend to not build units. I've also replayed turns where I got votes for Diplo win by building infrastructure and wealth but while building ICBMs and Tanks I lost votes from three AI's. While this makes sense on one level it is very frustrating that the AI can know things it shouldn't and that I can ruin myself in yet another unforeseen way.

Thanks for the reply TMIT

 

[Cusanus]

So in my last few games it quickly becomes apparent that I'm in a position to be boxed in with less land than I want. I can squeeze in 9-12 cities but the truth is only about 6-7 of them are on juicy sites without tons of desert/tundra tiles or so food poor as to be negligible pre-Biology.

You shouldn't even settle most of the bad ones, usually. At least not early.

 

[Calouste]

I'd be quite happy with 6-7 cities on juicy sites after the initial settling period. Heck, I'd be happy with 4-5 cities on juicy sites or 2-3 on juicy sites and another 3-4 on decent sites (1 food resource plus some river and/or hills).

 

[Mantic0re]

it just doesn't seem like enough when the AI with the fewest cities has managed double the land area and number of cities (around the time Liberalism is discovered). It seems like I've got my back against the wall and I don't like it.

 

[Calouste]

What kind of map size, script and difficulty do you play on? I play on normal fractal maps on monarch most of the time.

 

[pollock]

These dont sound like rush oppurtunitys. Can just "attack" with swordsmen and catapults or macemen and catapults/trebs.
Rushing generally involves going straight for the attack with everything geared for it. Expanding and then decididng to rush isnt really a rush in my opinion

 

[NihilZero]

Agreed with Pollock. It's important to have clear in your mind the distinction between a rush and later war, because the opening strategies involved are radically different. In a rush you build the minimum number of settlers and settle the minimum number of cities, and you gear everything towards attacking as early as possible. Quite what this magical minimum number is, is up to debate, but the point is that you want to catch the AI while it is in early expansion mode, and you do this by effectively sacrificing your own early expansion phase.

I've not seen anything written on HA but I tried them with 5 or 6 cities and 9 workers.

Vicawoo started a thread quite recently on HA rushes, you should be able to find it easily enough. Other options are chariots, and swordsmen to an extent, although swords can be tricky as they come later than axemen.

If you have room for six cities then a later attack is very feasible. A smaller empire techs faster, and if you don't have enough production then there's the draft, which is one of those key difference-makers in Civ that you need to exploit to win in some difficult circumstances. Don't get too hung up on number of cities, although it can be disconcerting to see the AI double or triple your city count. Sometimes one super-capital supported by a few less-than-stellar cities can be enough to pull out a win.