Managing your civ for the benefit of your UUs

[Theseus]

Originally posted by ducki
Another thing to consider - build a buttload of Regular Galleys before you build a harbor - these are your transports - then build a Harbor and crank out a bunch of Vet Galleys - this is your Escort Force - since they are all Vets, they are both the first to Defend and easy to identify. This could be risky.

GODDAMN, I never thought of this! WHATEVER your level of ships, the transports should be regs!

[ducki]

It can be dangerous, Theseus, like I said.
If you get a bad string with the RNG, then your transports are even more at risk, but if you have enough of an Escort, you can crank out all your "real" transports before you can afford to build the Harbor.

Not for everyone, due to the lower HP risk, but might save you quite a few turns earlier on.

[Catt]

I haven't tested it, but nbarclay posited the same theory (reg galleys as transports, vets as escorts), wondering if both were vets which would step up to defend. Soren posted to say that all else being equal, an empty galley would defend before a loaded galley. So the reg-vet combo might not be as important.

Catt

[Dominae]

Strangely enough, if you're being very aggressive early on, you won't have time to build Barracks, and so most of your force will be trained as Regulars. Having units in the field ASAP is far more important in my experience than waiting to make them Veteran. This is especially true for boats, where the chances they'll ever get into battle is slim. Of course, once you're industrial base is moderately strong, Barracks and Harbors are a great addition. People will probably disagree with me on this, but the main strength of the Militaristic civ is that it can send out Regular units and expect them to become Veteran with some well-planned skirmishes. Thus early Barracks aren't as necessary as they seem (despite their low cost). Non-Militaristic civs benefit far more from (expensive) Barracks, as their units are likely to stay "weak" for a while longer.

In AU 205, I'm playing the Vikings. I only built one Barracks early, and this town provided a constant supply of defenders (Spearmen then Pikemen). The other towns set to produce military units did so without Barracks, but my Archers, Swordsmen and whatnot were mostly Veteran or Elite by the time I got to Invention. At that point it was a simple matter to hurry some Barracks to keep my units up to date (incidentally, saving some gpt on Barrack upkeep).

I'll post some strats on the Vikings shortly...


Dominae

[DaveMcW]

Militaristic gets 50% Regular unit promotions, compared with 25% promotion rate for other civs. That's not enough to replace a barracks, in fact by giving up a barracks you start that much farther from an early elite/leader.

Here is the break-even point for the early units:


8 Reg Warriors = 24 HP for 80 shields
Cheap Barracks + 6 Vet Warriors = 24 HP for 80 shields

4 Reg Archers = 12 HP for 80 shields
Cheap Barracks + 3 Vet Archers = 12 HP for 80 shields

Cheap Barracks + 2 Reg Horsemen = 8 HP for 80 shields
3 Reg Horsemen = 9 HP for 90 shields

16 Reg Warriors = 48 HP for 160 shields
Expensive Barracks + 12 Vet Warriors = 48 HP for 160 shields

8 Reg Archers = 24 HP for 160 shields
Expensive Barracks + 6 Vet Archers = 24 HP for 160 shields

5 Reg Horsemen = 15 HP for 150 shields
Expensive Barracks + 4 Reg Horsemen = 16 HP for 160 shields


Conclusion: Militaristic civs should always take advantage of their cheap barracks, unless they are doing a kamikaze rush then turning into a builder. Non-militaristic civs take longer to benefit from barracks, but if a city is planning to spend 160 shields on military they should invest in one.

[Dominae]

Nice analysis DaveMcW. However, I wasn't arguing that Barracks aren't useful, given enough resources and time; you didn't take into account, among other things, transit time, Settler production and quality of defense. My point is that early tempo (unfortunately, an "intangible"), in my experience, isn't helped by Barracks when playing with Militaristic civs.

I'm perfectly willing to grant that my strategy here (early "tempo" rush) may not be as effective as the slower build-up of massive Veteran forces. To show this, you would have to demonstrate the cities I've conquered earlier aren't as "useful" as the same ones you conquer later (compounded over the number of turns in between, of course).

My Archer strategy that I've been using recently uses a "less is more" mentality. I'll usually just build enough Regular Archers to hamper the nearest AI. These shock tactics are great against the AI simply because Settler "bopping" works so well; the AI can't figure out what you're trying to do, so you can do more with fewer units than you could against a human opponent. Contrast this with massive Veteran Swordsmen or Horsemen "rushes": the extra time spend means your victory will be more decisive, but at the expense of raw speed. I don't think it's obvious which is better.


Dominae

[Dominae]

Even if I turn out to wrong about Barracks, the case against Harbors may still stand. Assuming you don't use early boats for naval offense (which I never do), all you need is the 2-unit transport capacity. The fact that Galleys have 3HP to the Barbarian version's 2 is nice, but not essential. In my games, the Barbarian ships aren't very good at tracking my ships. So the added strength of another HP is largely irrelevant, and consequently Galleys should be built before Harbors (unless you really need the food, or you can wait to put your boats into the water).


Dominae

[ducki]

Here's a brief one.

Carthaginians and Numidian Mercs:

In your despot days, particularly before you have "inner" cities and "outer" cities, Warriors and Archers(later Swordsmen) are your primary defense. (You don't want to blow a GA because the AI got froggy.)

Once you have either a)an inner and outer "ring" or b)Monarchy/Republic, start making these guys from a Barracks town, first posting them in interior cities, again, protecting against an unwanted GA - odds are you have Swordsmen by now anyways, use these as defenders in your outer/border cities.

When ready for a GA and prepping for war, rotate some Mercs to your borders and begin training replacements in case the AI comes at your core.
---End of Brief Version---

---Begin Ramble---
If you like, send 1 or two along with a Swordsman raiding party/invasion force to cause the GA. Or just use them as pillagers that will be very difficult to stop in the Ancient Age, and should easily take out left-over Warriors and Archers.

Otherwise, be happy with your premature pikemen and do not upgrade these guys until Replaceable Parts unless you have a war and need Muskets at the borders. This saves you a bundle of cash until your bankroll should be big enough to not care. Saving cash for Knight and Knight UU upgrades sure helps a cash-strapped Empire.

The most beautiful thing about Mercs(much like the other Ancient 3-defense UUs) is that you can keep these Pikeman Replacements around until Replaceable Parts, at least I've been able to in my Carthage games, with a few border town exceptions.


The trick is avoiding a premature GA, which is easy enough. Either take the battle to your enemy, or don't use them in border cities until you are out of Despotism. Having Swordsmen for defense, (especially if you don't just Fortify and Forget and are instead, a "proactive defender") is, perhaps, even better than having Spearmen.

[GusSmed]

Originally posted by Catt
My experience is that you have it backward. Gold is both more easily acquired and a more flexible resource than shields in the very early game.

I don't know what you're smoking, but in my experience, until you've got the techs you want to start a war, you're always playing catch up with the AI. Every coin you get is one you could potentially spend on research.

Sure, you can use the "40 turn rule" to your advantage, but if you do too much of that, you end up delaying your attack until the AI is strong and ready to counter, instead of catching them while they're still thinking about expansion.

Any river tile automatically produces one gold; a road does the same. Shields are available only with bonus grasslands or a mine.

You talk as if roads were free, or river tiles were more common than grasslands with shield bonuses. You have to build roads the same as you have to build mines. On some maps rivers are common, and on others they're quite rare.

You also get shields from plains. Provided you have a source of fresh water (always a big if), it's faster to get a 2 food / 1 shield / 1 trade square through road + irrigation than from a blank Grassland square through road + mining.

- Gus

[Theseus]

Gus, I'm not sure I follow you...

In the very early game, you can only effectively reseearch so much; everything else goes to gold (or maybe some luxury).

And regardless, there are many well-established early warfare strategies based upon achieving certain early enabling techs.

Maybe I just don;t get your point.

[FrustratedPoet]

Originally posted by GusSmed


I don't know what you're smoking, but in my experience, until you've got the techs you want to start a war, you're always playing catch up with the AI. Every coin you get is one you could potentially spend on research.

Sure, you can use the "40 turn rule" to your advantage, but if you do too much of that, you end up delaying your attack until the AI is strong and ready to counter, instead of catching them while they're still thinking about expansion.

Fair enough, but sometimes you don't want to research quickly. If you place emphasis on gold rather than production or research you can - for example - build lots of cheap warriors while waiting for your slow research to Iron Working, when you research the tech you will have enough gold to upgrade hordes of the Warriors to Swords and sweep your continent clean.

In the early game (if you're playing warmonger), you only need a low level of research development. If you get the timing right (easy with practice) then you can place a very low priority on Science up until you massacre and conquer neighbouring Civs and take all the techs they've so kindly researched for you.

You don't need to delay the attack until the enemy is ready for you. They certainly won't be ready for a huge Swordsman rush until they have Pikemen, and they'll be dead long before then.

Of course, if you want to play as a builder research is more important.

Also, if you play as an Industrious Civ the roads are built so quickly the almost are 'free'.

[BRC]

My experience is that you have it backward. Gold is both more easily acquired and a more flexible resource than shields in the very early game.

Gus... I think that this quote is right on. No offense. It's just that gold can be used in many different ways. Upgrades may seem like a waste of money to you, but this gives you the edge to either give yourself some room to grow, acquire some needed resources, or clear off your continent. You can't honestly tell me that you wouldn't trade 2000 gold for a continent after your horses upgrade. Using the luxury slider early in the game allows you to get your production level up. I wish I would have known about this tactic when I started playing. 50 gold could be the difference between finishing the Great Library first, or having to switch it over to the Great Wall. Having gold in reserve isn't a bad thing either. If you really need to investigate a city, or you need to throw in a little extra in a trade with the AI, gold is what's used. You can't use shields. Wealth stinks. All I'm saying is that gold can be turned into shields, but not the other way around.

Gold is a little more abundant, too. If you need more shields, you can lumberjack. That's about it. If you need more gold, you can find it somewhere. Trade techs for it if you need it bad. You won't get a lot, but if you need it... Barbarians have gold. Get it from them. Work tiles with roads. You can find gold if you really need to. I would rather be two techs behind with a fat bank than in the tech lead and barely hanging on to my cities.

[Catt]

Originally posted by GusSmed


I don't know what you're smoking, but in my experience, until you've got the techs you want to start a war, you're always playing catch up with the AI. Every coin you get is one you could potentially spend on research.

I'm not smoking anything - you could be a bit more civil.

Some players don't wait to acquire techs to start a war. A good warrior stomp or an archer war will do just fine. In any event, for those techs that greatly assist my typical ancient warfare (Horseback Riding and Iron Working), I want them to take awhile to research. I need that time to set up a core production center and to start building warriors and chariots for later upgrades. Playing the strategy as many commonly do, there is no "playing catch-up" -- you are behind, and you stay behind without any effort to catch up -- you will catch up through warfare and through later building.

I've found that spending aggressively on research early rarely pays a decent return on investment. Spending little on research but stockpiling gold tends to provide a better return.

Sure, you can use the "40 turn rule" to your advantage, but if you do too much of that, you end up delaying your attack until the AI is strong and ready to counter, instead of catching them while they're still thinking about expansion.

Again, there is no reason for you to "run out of time." A leaisurely research approach followed by mass upgrades and warfare, and you should always find a suitable victim. Except in very rare cases (for example on Deity with a lot of space between you and your nearest neighbor) you should never find your self charging pikemen with your horsemen.

In my experience, spending aggressively on research and then having to build 30-shield swordsmen causes a greater delay in "feeling ready" for war. Thirty shields (even 20 for a horseman) in the ancient age is tough to come by at the higher levels. You don't have the happiness infrastructure to support Cities (as opposed to towns) -- and without citizens, you have few shields. Gold, on the other hand, is generated a plenty even by a bunch of 2 or 3 population towns.

You talk as if roads were free, or river tiles were more common than grasslands with shield bonuses. You have to build roads the same as you have to build mines. On some maps rivers are common, and on others they're quite rare.

Actually, you must not have read my post. I most certainly did not imply roads were free -- I compared the worker investment in road-building directly to worker investment in mine building.

You seem to believe that shield acquisition is straightforward. I say it isn't. You need the terrain to cooperate and you still need to invest worker time to produce shields. In addition, you must invest twice as much (50% in some cases when industrious) worker time in building mines. The real "shield producers" -- mined hills -- are only infrequently attractive citizen tiles, since they don't produce enough food as needed to support a citizen (and absent a bonus resource, only flood plains will make up the shortfall until you get out of despotism).

Finally, note again that shield generation requires working citizens. Working citizens require happiness providers. With an early resource effort devoted towards gold, using the entertainment slider comes at less relative cost -- assuming you're playing the higher levels, shield production will require a decent gold investment as well.

You also get shields from plains. Provided you have a source of fresh water (always a big if), it's faster to get a 2 food / 1 shield / 1 trade square through road + irrigation than from a blank Grassland square through road + mining.

Correct. But beside the point. The issue is acquiring gold versus shields. If you want to deploy early workers efficiently and develop well-rounded tiles, irrigating and roading plains is clearly better than mining and roading grassland. My point was that roading -- and simply roading -- generates additional gold for the empire. Roads also serve both you expansion and your military mobility needs. It is not uncommon for me to set workers to building an extensive road network, including roads to the expected front, and then coming back to mine as needed. In the meantime, I am generating gold by having citizens working roaded tiles.

As a final point that I find important in comparing tactics, I often compare proposed actions to AI actions. Your approach of researching as quickly as possible and then building an elite force from scratch exactly mimicks the AI's approach. Playing at the higher levels, you're behind 2 eightballs with this approach -- (1) the AIs start with extra units; and (2) the AIs have production advantages -- so your 30-shield swordsmen cost the AI only 24 shields on Emperor and 18 shields on Deity. On the other hand, the "stockpile gold and mass-upgrade" tactic allows you to get out from behind each eightball to some extent. You build only "shield-cheap" units, eventually allowing you to offset both the AI's numerical advantage and production advantage. You sacrifice technology advancement for gold, but you employ gold to bring your technologically inferior force up to par in one fell swoop. This also has the advantage of allowing you to immediately field a force of XX horsemen or XX swordsmen when the AI opponent has only manually built several (in addition to the AI's inferior forces previously built but not yet upgraded). Finally, your sacrifice of early technological advancement is temporary, not permanent. You have secured for yourself (1) "cheaper" tech costs through tech devaluation, and (2) relative strength advantage which should allow some degree of tech extortion in the context of peace negotiations. You generally can't make up for lost gold-generating opportunities as the AI tends to spend it as quickly as it comes in. So, mimick the AI's approach or do something differently and exploit the difference.

You proposed an early-game tactic. I challenged the relative efficiency of the tactic. I tried to explain why an early focus on gold and an upgrade tactic is in most cases a better approach. I've tried to do so again. Maybe I haven't done a very good job at explaining.

But you haven't explained why maximizing research and then building 30-shield swordsmen is better than minimizing research, building 10-shield warriors, and then upgrading to swordsmen at 40 gold a pop. You haven't explained why building 10-shield chariots and mass-upgrading to horsemen at 20 gold a pop is less attractive. Your explanation(s) seems to be "you'll fall behind." That's not enough to support your tactical argument.

Catt

[ducki]

I understood it the first time, Catt, but the second explanation was even better, thanks!

P.S. I was pretty sure you weren't smoking anything.

[BRC]

Note to self: Get Catt a little riled up and get a great explanation.

I find that if you try to research in the ancient age (for the most part), then there is not as much use in the early extortion wars. Might as well save the gold, because you will get the techs anyway.

Gus- The "smoking" bit was a little out of line, in my opinion.