When a unit attacks, the combat result is determined by RNG. Say RNG decides that your attacker loses four out of four rounds. A horsemen have a good chance of retreating, and when it does it saves you 30 shields in cost of replacing it. After it heals, it's as good as new. On the other hand, a sword would have died because it will not retreat. You will lose a unit if the battle goes awry. In civ, you want to minimize risks and get the most out of what you've got. Horsemen's retreat ability gives you better chance at minimizing losses, as the risk of dying in combat is reduced.
I bring cats with my swords most of the time. If 3 veteran spearmen defends a town and you bring 10 swords and 30 cats. The 3 spearmen have a great chance of being red lined, if not seriously injured. Say 1 spearmen had two hit points left and the other 2 were red lined, it would switch to the 2 hit point spearman. Even if the spearman was full health, the swordsmen would have a greater chance of winning the battle. So, now that the spearman has only two hitpoints left, I will times 3 by 2. I timed 3 by 2 because the swordsmen has 3 attack, and the i timed it by 2 because the swordsmen has 4 hit points left, but I deducted 2, the amount of which the spearman had left.
That concludes to six. If I did the same with a horse, it would be 4. That's 2 more attack. Now, just to be fair, I'll include the retreat bonus, so... I think a vet horse has 60% retreat bonus. So, this horse would likely retreat, let's say it did. Btw, the horse is the attacker. the horse would go one square away, red lined. In this case, the city has three spearmen, so if it was me playing, I would send one after the horse. A spearmen has 1 attack. Time that by 3 because I deducted the one hitpoint the horse had left.
That leaves us with 3. The horse has only one defense, and he retreated to a grassland. So he has only 10% bonus, which is basically 1.1 defense for him.
So a 3 attack verse 1 attack, odds are the 3 is going to win. And if he doesn't I could just send a different troop from a different town (The troop would get their because towns are usually spaced like CXXC which means he has enough . Even if it was CXXXC, he would still make it in the right angle. And if you are at war, then you should be sending constant supplies of troops anyway, so you are more than likely in reach of the horseman. But that's only if the horseman survives, which would be slightly unlikely.
So, if the swordsman attacked, he would have a greater chance then the horseman, which means in the end swordsmen have a higher survival rate.
That assumes that:
1) They have a good road network in the direction you want
2) Said neighbour won't have the odd worker/unit/whatnot blocking a road tile, causing your movement to slow.
3) You want to sign a binding treaty that ensures peace with them for 20 turns, which I find undesirable
1) I wanted to say this in my other post but I didn't think I would need to. Anyway, if you play Regent-Monarch+ then the AI will have their roads developed much quicker. They would at least have all their cities connected by the time swords and horses are out (I play monarch, now).
2) Who said they'll only have one road connecting their cities? They will probably have more than a single road connecting their cities on Regent-Monarch+.
3) This doesn't necessarily mean by going through other AI civs, it could be your next door neighbor. But a ROP agreement isn't that bad. Most of the time the AI is willing to accept ROP deals without anything in exchange. And while the agreement is on, and you've finished your war with the enemy, why not pick on another neighbor? I don't find that too undesirable.
Furthermore, even if the enemy is next to you, horsemen still have an advantage over swords; speed. In Civ speed kills. Speed allows you to have:
1) Flexibility: You can quickly shift your troops from one front to the other
2) Faster campaign: AIs sux at defending. Horsemen can quickly take one city and then move to the next target twice as fast as slow units i.e. swords. Since AIs don't understand the advantage of speed. their slower units take more time to reinforce the cities. With slow units, you're at a disadvantage as your attackers are moving as slow or slower than the enemy (you can't use the roads in their territory).
3) Surprise: With horses, you can declare war and raze at least one or two outlying towns on the same turn. This is because horsemen can attack a town from outside it's cultural borders. Proper planning can help you raze a good number of towns before any sizable counterattack shows up.
1) But if the horse has already attacked and retreated, he has no cannot run away to any other troops attacking him. And, horsemen are
slower than swordsmen in your own territory. But in open land, or were the enemy horseman comes from, yes they are faster. Still not good enough to make me want to sacrifice 1 attack and one defense.
2)What about the catapults? swordsmen are as fast as catapults, so they don't have to wait. And if you use only horsemen and neglect catapults, then the swordsmen are stronger, because catapults are helping them. They aren't with horses if they just run off and attack cities by themselves.
3) Again, the same thing.
On the contrary, I do not think that is good. If my unit is losing the combat, the least it could do for me is save itself to heal and fight another day. When swords die, the 30 shields in cost is lost. Granted, killing enemy units may be considered compensation. But, why throw away your units? The retreat ability keeps losses lower, as I mentioned above. Whenever my horseman retreats, I would know that if in that particular combat I had used a swordsman, my swordsman would have died without question. By using a horseman, I use up the "bad RNG" combat without losing anything.
I answered this already...
What about the catapults? swordsmen are as fast as catapults, so they don't have to wait. And if you use only horsemen and neglect catapults, then the swordsmen are stronger, because catapults are helping them. They aren't with horses if they just run off and attack cities by themselves.
And swords don't have double attack compared to horsemen
Swords may kill more, but they die more too. I'd rather have a lot of retreats with horsemen while killing a lot with very few losses, than lots of win-or-die combats with swords that kills as many units but with more deaths on my part.
I strongly disagree. In the long run, archers sux. Why do I say this? Several reasons:
Archers DO NOT suck, IMO their better than horsemen. Anyway, this isn't so much about the archers this is about the swordsmen, and swordsmen are so much better than HM (Horsemen) that it is indescribable.
1) Pre-emptive strikes?
-That works only on defense. A good, effective campaign means you must not let the AI do a lot of the attacking (if any). The human players are superior to the AI in terms of managing combat and manipulating variables that determine odds of success in battle. It's a shame not to use this advantage we have.
HM are no threat to me, apart from the pillaging, but I can just kill them to solve that problem.
Anyway, if the AI HM attacks and retreats, I'll at least have on troop to kill it, I've said this already above.
2) Upgrade path
-Horses upgrade to knights. Archers upgrade to longbowmen. Who needs longbowmen anyways when you have knights? Same attack rating but knights are faster. Often, having a good number of knights before anyone else can be a gamebreaker. This is especially so with powerful knight UUs such as the Chinese Rider and the Arabian Ansar Warrior. These two have three movement points. That's enormous!
I admit, knights are good, but no. I have to concentrate on the present, not the future, and in the present swords and archers are better. And I wouldn't build archers if I had swords, the main topic here is the sword, not the archer.
And swords have a pretty good update path too. With archers, I just stop producing them, and they slowly die off in battle. So, the longbowmen upgrade path is no concern to me.
Some may argue over the costs between knights and longbows. Knights are almost twice the cost in shields, yes. But good players know not to build up their deadly host of knights entirely from scratch. A powerful tactic ( for me, at least) is to build a lot of horsemen and accumulate tons of cash before Chivalry comes in. In some of my games, my instant-40 knights the turn after I research Chivalry handed me the game. Trust me when I say that instant 40 Longbowmen in the same scenario will not have the same effect
Knights are the first, really dominant mobile unit, so I would put knights on a higher priority, thus I build knights instead of Longbowmen. But I don't build horsemen to upgrade them to knights, I don't build horsemen at all.
My advice would be to test it out in your own games. See what works for you, and compare the plus and minus of each. Or, you could read up some games by great civ players, and see their reasoning for their choice between horsemen and swords.
Btw, if anything I say offends you, let me know, because I didn't mean anything bad with what I say, I certainly don't want an argument in this thread.
If you study carefully the benefits of a 2 point movement
and retreat, you get to the conclusion that Horsemen are
factually better than Swordsmen, the only disadvantage being that you can't use them for defense, which you shouldn't do with swordsmen either. I remember someone a long time ago posting some exact formulas proving this. And he was not anyone
(I don't remember who, but a great HoF player in any case). That being said, Horsemen are my favorite AA unit, except UUs.
IMHO, Horsemen are undoubtedly better than swordsmen when playing on the lower levels. There is no question about it.
I strongly disagree.
From what I know about the way Aceman enjoys playing, I would guess that he is often spending time building infrastructure while he could be building horses.
It's more, building swordsmen and cats, and not building horses.