How many units should I put in my SoD?

[Montymolethedog]

More specifically, how many units relative to the amount of defenders in the city being attacked shall I put in my SoD?
And how many units of each kind? (eg. seige, attack, stack defense, medics?)
Thanks.

 

[elitetroops]

Depends on what units you are talking about. What are the defenders, what units are you attacking with? The stronger your attackers are relative to the defenders, the less units you need. Also depends on unit types. Attacking with mounted, then probably no siege. Attacking with melee, then probably there should be siege. How much depends on defenders, cultural defenses, walls/castles and so on.

 

[Enyavar]

Half as many attackers as defenders. Your properly promoted tanks will gun down two AI archers in a single round.

No, really, I also could find use for an update of this comprehensive guide here. A link to such a dissertation will suffice, thanks in advance.

 

[Gumbolt]

For immortal. The stack should be strong enough for it's purpose. Early HA rush might contain 8-12HA. (1000-1200bc) Axe rush a similar amount. 9-10. Phants pult rushes. 15-20+ units. Cuirs rushes 20+ units? Also depends on the size of the Ai stack. If the Ai has a 30+ strong stack you know you need more units.

It's all relative. On warlord 2 warriors early on is enough to rush an Ai capital. Where on immortal you would need 15+ warriors to even have a chance. Hence why it is mainly done on lower levels.

Overall more info would help. Difficulty level. Maybe even a save. Of course all the above suggestions also depend on year you do your rush. A rush in 2000bc needs less units than a 1500ad rush.

 

[Tristan_C]

There's no rule about it. The more units, the more options. If you "overbuild," it means you can cap a city and move right along to the next, leaving injured behind to heal and catch up later. If you just barely manage to capture, then you have to hold the position and reinforce before advancing. No hard decisions here really

As for proportions (when not solid mounted), I like to start off at 2/3 siege and keep building siege during the war. They do bombardment and they take the losses. The mop up units, melees or rifles generally, have an easy life and shouldn't have to be replaced much.

 

[Chakasulu]

Hello all. I have a couple questions about the use of siege I was planning to ask before seeing this thread, then figured it would be be best to piggy back on since they are very much related to what the OP was asking. I play at the Emperor level.

@elitetroops: Is there any reason other than speed why you wouldn't normally bring siege with mounted units? Specifically Cuirs, since HA and chariots come too early for it to be relevant and elephants come as a packaged deal tech-wise and with a speed of 1 there is no strong advantage for ignoring catapults.

Is it because in addition to ignoring walls/castles that it is assumed that Cuir strength is high enough vs. the opponent's units that speed has greater utility than that of collateral damage/bombardment?

@Enyavar: I read what Sisiutil wrote in 2006 about catapults and I was curious to see if current players feel his advice still holds true or may be considered overly conservative, which is how I generally play and wonder if I'm needlessly slowing myself down:

"Ideally you should have four Catapults for bombarding away a city’s defenses in one turn ... These bombardment siege units should never be used to attack the city. So you should also have some “suicide” Catapults that will perform the initial attack and cause collateral damage. A good ratio is 1 suicide siege unit for every 3 city defenders" - Sisiutil

 

[Gumbolt]

Why do cuirs need seige???? Because if your only facing Pike, mace, LB and knights they are more than capable of handling these. Remember trebs/catapults vs castles would need an army of seige units. It's only when you get to rifles that you really need seige. I think you hit the nail on the head for the other advantage of cuirs. The big plus is the 2MP which enables you to race across Ai land. 1MP units mean he whips 1-2 extra defenders.

Cuirs are more than capable of taking down cities often with odds of 30-70%.

 

[Chakasulu]

I see. So I think my weakness is I'm not aggressive enough. If I saw odds below 70%, I'd think "oh no way, I need to soften that up a bit," leading me to believe I need siege. In reality if instead of siege I just whipped more cuirs it sounds like I'd be just fine.

Thanks Gumbolt.

 

[Gumbolt]

It's when the odds are below 10% you worry. The reality is after first 2-3 attacks the odds can only improve. I always want 2x attackers compared to defenders unless the starting odds are above 70%.

 

[elitetroops]

@Chakasulu: Yes, with mounted I meant HAs, Cuirs and other 2 move units. Elephants of course usually go with pults, even if they are technically mounted units.

In a HA war you should be prepared to attack at very low odds if necessary. Most of the cities will be easy to take, but every now and then there can be tough walled hill cities, or worst case scenario someone gets Chichen Itza in their hill capital with 60% cultural defenses. In that case I'd try to use spy revolt if possible. But if not, then just expect to lose up to 2 units/defender and you should be fine. With HAs, even if the first attacks are at <1% victory odds, you still rarely lose more than 2 units/defender (assuming the defenders are classical units).

Spears should be avoided in cities with HAs. Most of the time you can lure them out. If there's a HA in range for them to attack, they will mostly do it. Best is if you can place a C1+Shock HA 2 tiles away from their city, on flatland, so that they can attack only across a river. They still often will attack, despite odds not being at all very favorable for the spear. If they win, you get to take out the severely damaged spear in the open with no defensive bonuses.

 

[Chakasulu]

@Chakasulu:

In a HA war you should be prepared to attack at very low odds if necessary. Most of the cities will be easy to take, but every now and then there can be tough walled hill cities, or worst case scenario someone gets Chichen Itza in their hill capital with 60% cultural defenses. In that case I'd try to use spy revolt if possible. But if not, then just expect to lose up to 2 units/defender and you should be fine. With HAs, even if the first attacks are at <1% victory odds, you still rarely lose more than 2 units/defender (assuming the defenders are classical units).

Even more evidence that clearly, I need to grow a pair. Elitetroops and Gumbolt; I thank you.

@Chakasulu: Spears should be avoided in cities with HAs. Most of the time you can lure them out. If there's a HA in range for them to attack, they will mostly do it. Best is if you can place a C1+Shock HA 2 tiles away from their city, on flatland, so that they can attack only across a river. They still often will attack, despite odds not being at all very favorable for the spear. If they win, you get to take out the severely damaged spear in the open with no defensive bonuses.

That kind of tactical approach is something lacking in my games. I will start looking for opportunities like that

 

[Lexicus]

It all depends on the time frame. If your targets get grenadiers or curs of their own you'll probably find it profitable to wait for cannons to pull of your cur attack.

Once I was fortunate enough to attack Lincoln with Cavalry when his strongest defenders were longbows. That was a glorious war - big stack of cavalry was attacking his 60 and 80% defense cities at 70%+ odds.

Worth mentioning, key when doing HA rushes in particular is getting to target AI's core and capital as soon as possible. These are cities that likely can produce close to 1 unit/turn so the faster you capture them the fewer units you'll need to deal with. Peripheral cities can then be mopped up at leisure.

 

[6K Man]

Pre-siege:
Axes: 2.5x the Archers in AI Capital. 3x the Axes.
Chariots: 3x the Archers in AI Capital. 2x the Axes.
Horse Archers: 2.5x the Spears in AI Capital. 1.5-2x anything else

One should have Medic 1 promo.

Post-siege (Catapults or better):
At least as much non-siege as there are defenders, plus 2. 1 super medic (anything with Woody3 + Medic promos) or 2 regular medics (Medic 1, 2, or 3). 1 stack defender (spears/pikes if facing elephants/knights, for example) per three units they counter. And as much siege as all your other units combined.

 

[Centimus]

Sorry for the sidetrack, but I think siege is the first aspect of the game that calls for modification. The whole idea of the attacking siege units being drawn into a duel with their targets, rather than simply bombarding them from a safe distance, is just ridiculous IMO. So I've given all siege units a high withdrawal probability, which means suicide attacks become rather unusual. At first I worried that it would drastically alter the balance of the game, and the proportions of AI armies, but I think it works out pretty well. I'll happily share some advice if such a modification sounds interesting to you

 

[elitetroops]

Sorry for the sidetrack, but I think siege is the first aspect of the game that calls for modification. The whole idea of the attacking siege units being drawn into a duel with their targets, rather than simply bombarding them from a safe distance, is just ridiculous IMO. So I've given all siege units a high withdrawal probability, which means suicide attacks become rather unusual. At first I worried that it would drastically alter the balance of the game, and the proportions of AI armies, but I think it works out pretty well. I'll happily share some advice if such a modification sounds interesting to you

If this doesn't alter the balance in your games, you are not using siege right. Siege units are already overpowered. Increasing siege withdrawal probabilities would allow you to take most cities without any losses at any time, which would make offensive warfare way too easy.

 

[elmurcis]

Last week I managed to Lib Steel before any guy got even Gunpowder.. With 18 Cannons and ~18 random medieval units (just to be able to kill enough units/turn to forward) I killed 2 AI before I got to Grenadiers, 1 more with Grenadiers (had CR2 Maces) and 1 more with Rifles (from Muskets) before any of Ai managed to get Curaissers out (yeah, but Curaissers<Rifles so I could move forward anyway). Losing just 2 Cannons and 2 WElephants meanwhile while attacking multiple cities same time and moving forward with all units (any defense whip from new cities). With CR3 Cannon odds (after earned some XP from few battles) there was no chance for any AI to resist this attack
Only "weak" period of siege is when Ai got Longs but player don't have Engineering yet.. well, than I don't expect any catapult to survive at all (in real world battles siege also was left behind if attack failed... crew could escape but not catapult/treb by itself)

 

[Centimus]

If this doesn't alter the balance in your games, you are not using siege right. Siege units are already overpowered. Increasing siege withdrawal probabilities would allow you to take most cities without any losses at any time, which would make offensive warfare way too easy.

I said it didn't *drastically* alter the balance, and I rarely take cities without suffering losses. Bombarding a city's cultural defence/units, and then waiting for my units to heal also takes a long time, as the siege units typically get badly damaged. I'm not denying this is preferable to suicide attacks, but I've also run into some serious trouble when facing huge AI invasion armies, in no small part because the same advantage applies to them. Btw, the cannon is the latest siege unit I put to extensive use - in the late game I rely almost exclusively on air support and Guided Missiles.

 

[gaash2]

Depends on how big the mismatch is and how patient you are, what time period, difficulty level etc. Often times I go with smaller than normal stacks and piece meal the enemy to death via invading 1-2 cities, extorting for peace, then rinse-repeat.