[BTS] Ancient-era rush at Emperor difficulty

[Oaq]

I wish to rush my neighbor during the ancient era at Emperor difficulty, standard settings, random leader—especially if my random leader happens to be aggressive. I have tried many times. Have usually failed. Why am I failing, please?

I seem to depend on blind luck. If my target can find no strategic resource (Horse or Copper or, depending on timing, possibly Iron) then my rush succeeds; but about 80 percent of the time, my target can indeed find a strategic resource, and about 70 percent of the time, my target can find the right strategic resource to counter my attack.

* * *​

The above sufficiently summarizes my question, so you can stop reading at this point if you wish. Otherwise, more details and an example follow.

Regarding my problem generally, combined arms in the attack would solve the problem, but for an ancient-era rush, it just seems to take too long to develop combined arms—for, even if my combined-arms attack succeeds, it's so slow that it gives a third power in the neighborhood time to eclipse me.

To give a specific example, during my last attempt, as Genghis Khan, I

• trained one Worker and two Warriors,
• researched Animal Husbandry,
• discovered a nearby Horse,
• founded a second city immediately next to the Horse,
• researched ancient techs the map made immediately necessary (on this map, Masonry, Mysticism and Agriculture),
• skipped ancient techs the map did not make immediately necessary (such as, on this map, Mining and Bronze Working),
• built two Barracks,
  
• researched the techs necessary to build Genghis Khan's unique unit (Horseback Riding and Archery, in that order),
• built two Gers (Stables),
  
• started training Keshiks (Horse Archers) as fast as possible, and
  
• started researching directly toward Mathematics to improve the chopping of forests.

I see no practical way to have proceeded any faster. Have I missed something?

During the Genghis Khan attempt, pictured, my target (Hammurabi) linked a source of Copper, located in a dark rear area I could not see, about the time I finished training my second Keshik. Therefore, when I launched my invasion, my target had new Spearmen to thwart it. I had had no time to link Copper of my own, hence no Axemen, so my invasion collapsed.

Indeed, in this instance, my invasion collapsed before I had even had time to chop any forests, for my lone Worker had been busy just developing necessary resources and building necessary roads.

Details vary from attempt to attempt depending on the random leader and the map, but most attempts similarly fail. What am I doing wrong, please?

 

[Zavior]

I think not going for mining/bw is a huge mistake already. With chops/whip you could have double the amount of keshiks you have. Why did you need masonry/myst here?

If you have the t0 save, please attach it!

 

[Oaq]

The advice is appreciated. To answer your questions, I researched masonry because Marble is found north of my capital; I researched Mysticism to let my second city access nearby Flood Plains; I skipped Mining because there was no hill immediately to mine; I skipped Bronze Working because Keshiks don't need it and, as you see, I was in a hurry. Perhaps wrong, but those were my reasons.

Per your suggestion, I have attached to my original post the turn-0 save.

 

[sampsa]

I can point out some mistakes even without seeing the save.

• two cities is worse than three, especially for an IMP leader. For example pigs+gold -spot would be very useful for your empire
• you should go mining-BW especially with that amount of forest and chop all the forest without math. Myst and masonry are useless for the rush even if you have stone/marble
• Gers are slowing you down a LOT. Would you rather have 4 highly promoted units or 8 barracked units? Latter will cut through emperor AI ~T80 while the former likely won't conquer many cities

So to summarize: you are doing a lot of things that slow you down likely because you think it will make the attack stronger. In reality the thing that gives most strength to the rush is SPEED.

 

[dankok8]

Skipping Mining/BW is a big mistake because chopping/whipping will lead to very fast production. You don't need Mysticism and Masonry if you're gonna rush with Keshiks. It slows you down and there is a ton of time to connect the Marble later. Always have food in the first ring anyways because even chopping a monument then waiting 10 turns for borders to pop is very suboptimal. With food in the first ring, city can grow right away. And yes 3 cities or even 4 would give you better Keshik production.

I recommend restarting with all the tips in mind. With rushes, it's ok to repeat till you get it right.

 

[Fippy]

I looked and not settling spots like fish + deer + double furs 3 tiles from your cap qualifies as sin
Adding that to other comments, you basically missed everything. Might as well be honest, it's bad.

 

[drewisfat]

1 worker 2 cities should really have been 3 workers 3 cities, grabbing BW relatively early (you have the wheel so worker won't be idle). A couple chops into faster worker/settler production, and then prechop the rest while teching HBR.

Metal doesn't end a HA rush. The AI doesn't know how to mass spears. It will also build axes with copper, which may be a less hammer efficient defense for it than archers / walls.

 

[Oaq]

Metal doesn't end a HA rush.

It doesn't? I had not realized.

The AI doesn't know how to mass spears.

Interesting.

It will also build axes with copper, which may be a less hammer efficient defense for it than archers / walls.

Even more interesting.

 

[5tephen]

@Oaq
By mentioning combined arms and axemen are you suggesting using axemen and horse archers together?
The general principle of 2 move units is they lose their advantage when paired with one move units.

 

[dankok8]

I doesn't? I had not realized.

Interesting.

Even more interesting.

Yea. AI is quite dumb to mass a counter unit. They'll just spam a bit of everything unless... the counter unit happens to be their UU. Then they will spam it. Keep that in mind before rushing Zulu or Maya with mounted.

 

[drewisfat]

I mean it's great if he has no metal because then HAs or keshiks can run around choking with complete impunity. But that's rare. I always assume metal units and it's not a deal breaker, even against an enemy with agg trait like Hammurabi. The reason the rush failed is entirely because you have 4 HAs instead of 12.

Axes would not help you here, that's just not the result of civ 4's combat system. When you attack a city with axes, they won't target the spear first, they'll target the archers or enemy axes. As a stack defender, not only are they slowing your stack from 2 movers to 1 movers, but it's literally not helpful. If you have an axe covering your horses, the AI just won't attack the stack, it will keep the spearmen inside the cities. If you don't have a stack defender, a spear may attack and most likely kill your horse, but now it's hurt and out in the open with no (or fewer) defensive bonuses and is easily mopped up by a second horse unit. That's the best trade you can get. It beats the spearmen camping cities where a single fortified spear might kill multiple horses.

 

[pi-r8]

Isn't the title a contradiction here? Horse Archers aren't an ancient-era unit, they're classical era. You'll almost never be able to research Horseback Riding before they research hunting and bronze working. A real ancient era rush is warriors, chariots, and axes.

Like other people said though, the trick with Horse archers is that they're so powerful they can still smash through spears, you just have to attack with 2:1 numbers.

 

[Oaq]

In the sample game, I devoted the first 15 turns to training a Worker. One would prefer not to have to do that. Is there a better way?

If I could somehow have saved the 15 turns, even my little band of four Keshiks would have sufficed to break the foe.

Unfortunately, I know how to save the 15 turns only when playing Huayna Capac.

Isn't the title a contradiction here? Horse Archers aren't an ancient-era unit, they're classical era.

Correction accepted.

 

[sampsa]

Well, you can always try to steal workers. However, normal play is to nearly always start with a worker. Many players can consistently win on deity by always starting with a worker and emperor is quite a significant step down in difficulty.

I think you need to get the basics into place as that is where the problems lie. I also think this is a tough map due to the horrible capital (one of the weakest I've ever seen) and very unsuitable for HA-rushing.

I'm playing a few more turns and then reporting...

 

[sampsa]

T82

Spoiler :

Worker first, mining-BW. Capital has extremely weak food, so it's better to just remain small at the beginning and leverage IMP to expand (well, or to move initial settler, but I wanted to play similarly to you). Mined the sheep, went settler right after worker. Chop. North first of course, it's a great spot. Chose to claim the western horse with the 4th even if it's very uncomfortable - need a monument. Monuments are horrible buildings (need the tech, then fast 30, then wait 10T...). It's much better to get the border pop from religion, library or caste. Just put to the 1st ring and you won't need border pops. Went for the 5th city as I thought I have a chance for pigs+gold, but lost it. No great spot, but the wine spot is at least insta-connected and has forests and river tiles.

5 workers, now chopping it all, whipping when running out of good tiles (means capital and fish-deer-fur will remain at 6-7, others whip immediately at size 4.) All cities have granary+barracks. 5 keshiks ready to declare next turn if Dur-K is not reinforced. 5 cities produce keshiks very swiftly so I think I can kill all met AIs with them.

 

[Oaq]

@sampsa, three questions, please.

1. Why Barracks but no Ger (Stable)?
2. Why go for the northwestern Horse when the southeastern Horse lies near a river?
3. If your game were continued (though I do not ask you to continue it), would Construction and its Catapults halt your Keshik-led (Horse Archer-led) conquests? Or would you expect to fight through even Construction and Catapults?

 

[Lennier]

I can point out some mistakes even without seeing the save.

• Gers are slowing you down a LOT. Would you rather have 4 highly promoted units or 8 barracked units? Latter will cut through emperor AI ~T80 while the former likely won't conquer many cities

So to summarize: you are doing a lot of things that slow you down likely because you think it will make the attack stronger. In reality the thing that gives most strength to the rush is SPEED.

@sampsa, three questions, please.

1. Why Barracks but no Ger (Stable)?

As for #1, see sampsa's reply. You can build the barracks while waiting for HBR; once you have HBR you should pump out Keshiks as fast as possible (if rushing.)

 

[rah]

most everyone has already said the major issues, but don't forget that getting BW earlier allows you to scout and figure where their metal is to pillage it faster. Especially with your unique unit's speed.

 

[Fippy]

+ barracks are only 25h with Genghis, very cheap.
Gers are a special case..estimate how many Keshiks you will get from a city quickly (the really important word here ) and you can find out if they are wanted.

Some examples..copper city with 15 base hammers and 8 forests to chop into Keshiks = sure a Ger here will be nice.
City with only 3 forests, one good food but no base hammers = should put all it's (whipping) effort into increasing numbers instead.

I don't agree with "never build stables or Gers for HAs", but peoples want to point out that building your stack fast has priority.
If you can do so while also getting 1 Ger in your best city that's nice, but train basics first.

 

[SittinDown]

@oag - it can be hard to get all this advice at once, I was in your spot not that long ago. if you stick with it - it will start to all make sense.

Something else not mentioned - you don't need so many roads (or any) - you have Keshiks. Their only specialty is ignoring terrain movement costs. Take advantage of that, and have your workers do other things like chop or improve tiles.