What's the trick to upgrading?

[ybrevo]

"Vel" - thanks for your detailed comments on upgrading. There was quite a few things I never thought of before.

Normally I upgrade my veteran troops (or is this an old Civ. expression?) and disband some with lesser experience. But I always keep some money in the bank for an upgrade that can turn the tiles in the very next move.

An example: I have a caravel protecting my fishing boats, and it is only *. Then comes a ** rival caravel from a nice lady I happen to be in war with. Odds are very bad about attacking it, and it will destroy my ship and fish in the next 2 turns. So instead I upgrade to frigate (just got oil a few turns before), her caravel tries to run away but is easily detected and part of the past.

My point is: Save something for an upgrade that really makes a difference (and why didn´t I put a frigate there in the first place ;-)

Ybrevo

[wodan11]

To me, upgrading is useful in only limited situations:
-- emergencies (of course)
-- speed (it's faster to upgrade than to build anew, and the longer you wait, the less time before your opponent gets new techs or, if you're already at war, whips or builds countering forces)
-- to preserve upgradeable promotions (e.g., Berserk amphibious)

In other situations, to me it's generally preferable to build new and disband the old. That's because I prefer to have my commerce going to research rather than gold. That assumes that the amount saved is more than I would get if I had that city building Research. (But that's an easy calculation to do.)

These days I also "save" a few experienced troops, sending them back to be a garrison unit or something. When I get a great general, I join him to one of those guys for a free upgrade, plus it gives a solid 50+XP right away. Makes my warlords MUCH less likely to die in combat.

Wodan

[Velociryx]

More thoughts on this topic:

I would refrain from ever disbanding old troops, unless you've so grossly overbuilt that you are crippling your economy, and even then...go pick a fight with someone and convert them into a couple of new cities, rather than disband.

I admit to occassionally disbanding excess workers, usually right after completing my extensive rail network, but even once all the real work is done, I'll keep a couple fistfuls around for revamping newly conquered territory to my liking, and/or repairing infrastructure that gets ripped up during fights with rivals. Still....8-10 (perhaps a few more if I've got a far flung, island or continent spanning thing going on) workers in the late game is plenty for the cosmetic fixes that are generally required.

Old, outdated troops though, are always useful.

There are a couple of different schools of thoughts here, and experimentation is really the only way you're going to get comfy enough with it to find your own path that meshes perfectly with your style of play, but I would say this:

If you always "build new" rather than upgrade, then you're not using your hammer counts optimally (by constantly having to build new troops when you've got perfectly good, upgradeable troops you've already invested in, you're building units when you could be focused on infrastructure).

On the other hand, if you try to upgrade EVERYTHING, all the time to the latest "top of the line" configuration, then you'll bleed money out horribly, which is also less than optimal, so find a balance (tip: it's less expensive to upgrade from a warrior to a rifleman than it is to upgrade from a warrior, to an axe, to a mace, and then to a rifleman).

Stuff you could do with them would include:

* Keep those old, outdated troops around, and as your empire expands, you can use them as "safe" interior garrison forces.

* You can also use them to stack in a newly conquered city to counter cultural flip (again, the culture flip formula only cares about how many troops you've got in yonder city....it makes no difference if they're warriors or grenadiers).

* In a pinch, you can sell tech, your world map, or resort to good old fashioned extortion to raise funds to upgrade a select few of these to bolster defense in a key area.

* In an invasion, land a bunch of them on a new front to cause a diversion, or, mix them in with your main attack force and use them to pillage enemy lands. Remember, you need not capture cities to wreck an enemy empire. If you take a couple cities on the border and then strip out every farm and cottage yonder civ has....how much of a threat do you think he's going to be to you in 200 turns time? Answer: None. You will have permanently de-fanged him.

You can also keep a close eye out for a strategic opportunity to mass upgrade...example...if you're the first person to rifling, you have a window of opportunity...a brief span of time when a mass-upgrade campaign will give you a VASTLY more powerful army than any of your neighbors.

By stripping garrisons down to a bare minimum in all your cities, and congregating a large force on the border, stalling tech long enough to upgrade all these troops to riflemen, you can *easily* double, if not triple the size of your holdings before your opponents can respond with a comparable force.

Is that worth a few turns of lost research to spend money on some upgrades?

You bet it is.

Another good time to do it is right after you research steel. All those veteran Cats and Trebs you have lying around....bang. A massive infusion of power that can *wreck* whatever rival you set your sights on.

So there are times when it's more potent than others, and those are the occassions when you really want to take it to extremes.

-=Vel=-

[Velociryx]

Another trick that I personally use with some frequency....

So let's say you get rifling, or some key tech, and you don't want to wait to upgrade your ENTIRE army before starting a war...so don't.

Upgrade enough to take a city or two, and bring your older guys along for the ride.

Take out a city...it becomes "your territory" (and you get conquest money).

Use conquest money plus your gpt income to upgrade a few more of your invasion force, such that your invasion is actually paying for the upgrade cycle. After all, the troops are now sitting in "your territory" (the city you just conquered), so they're valid targets for upgrading....

Nothing says you have to go whole hog before you start the fight...

-=Vel=-

[Yosho]

Eh...I don't usually disband workers. If you've got a lot of workers hanging around, you can usually figure out something useful to do with them, like "build backup road/rail networds" or "build a fort in a border square you cna't work" or "constantly micro-manage city squares" or whatever.

But, yeah, I upgrade a lot. Especally when you first get a new tech your neigbor dosn't have; the difference between having 10 infantry to attack his rifles NOW rather then LATER is big.

[Velociryx]

Oh, I don't always disband them...I have other nefarious uses for them in general, but I HAVE disbanded them before, mostly to shave a smidge on maintenance costs (by the time you finish your rail line, you have to ask yourself what you're gaining by keeping 40-odd workers hovering around (this, including captures from other empires, etc).

So sometimes....disbanding saves some coin, but more amusing than that is using them as bait for enemy troops.

The scenario works like this:

AI's can't resist undefended workers, so in battle, that means that you can arrange to know when and where enemy troops will be.

And...if you don't mind losing a couple workers anyways, then it's a cheap way to guide and direct enemy troops right into your kill zone....

A cheap trick?

Yep.

But entertaining.

-=Vel=-

[Yosho]

Oh, yeah, I've done that before too. They make great bait, especally if you're outnumbered and want to whittle down the enemy a bit.

Un-upgraded units can be used for that as well, although there you are giving the enemy experence and possibly getting him a little closer to a great general. Still, if I can trade one of my warriors for one of his tanks, I count that as a win.

Un-upgraded units, especally fast ones, are good for pilliaging as well; keep harrassing him on a secondary front by destroying his towns and mines with your cavlery or knights, and he'll send some of his units out there to deal with them, splitting his army away from your main stacks.

[wodan11]

Originally posted by Velociryx
More thoughts on this topic:

I would refrain from ever disbanding old troops, unless you've so grossly overbuilt that you are crippling your economy,

That implies it was a bad idea to begin with. Sometimes, sure, especially if you're ignoring all infrastructure to crank out units for an early war. But, in other situations, what usually happens (and is what we're talking about) is that you've got a "normal" amount of troops. Which I would define as = to how many your economy permits you without added maintenance. Then, you get techs and those troops become outdated.

and even then...go pick a fight with someone and convert them into a couple of new cities, rather than disband.

Easier said than done. If a handful of outdated troops could so easily conquer a few cities (assuming your economy can support those cities, in the first place, plus that your diplomatic situation has a convenient enemy on your borders), then there would be no point in upgrading at all.

If you always "build new" rather than upgrade, then you're not using your hammer counts optimally (by constantly having to build new troops when you've got perfectly good, upgradeable troops you've already invested in, you're building units when you could be focused on infrastructure).

Agreed, that disbanding is "wasting" hammers that you earned and spent on building those troops hundreds of years ago.

What you're arguing for here is throwing good money after bad.

No, that's not quite right. By saying "perfectly good, upgradeable troops" that implies that upgrading them is always a good idea (which contradicts the other statement that you could simply throw them at an enemy without upgrading).

What I would disagree with primarily is the implied assumption that there is no decision to be made... that the cost of upgrading is always less than the cost of building new.

If/when this assumption is NOT true, then we get into "throwing good money after bad" situations.

Stuff you could do with them would include:

Great tips.

One thing you don't mention though are that sending them after an enemy, to pillage, or to "easily" conquer a few cities, is that this basically gives the enemy a slew of free XP, wiping out those outdated units.

Wodan

[Jaybe]

* You can also use them to stack in a newly conquered city to counter cultural flip (again, the culture flip formula only cares about how many troops you've got in yonder city....it makes no difference if they're warriors or grenadiers).

According to CIV4UnitInfos.xml, this is not true. A Warrior has an iCultureGarrison of 3, a Swordsman's is 5 and for a Rifleman it is 8. Modern Armor has a iCultureGarrison of 16.

I have also tested this with comparing obsolete and contemporary units. The contemporaries definitely improve the city's revolt/flipping resistance.

[Verrucosus]

Another subtle rule I didn't know about. Thanks, Jaybe. (Though it was amusing to picture the citizens of a conquered industrial city being intimidated by Vel's warriors patrolling the streets with their clubs.)

[Yosho]

If you're despreate, obsolete units are also useful for defense; just because attacking units can only kill 1 guy per turn (well, attacking units without blitz at least), there have been times when I've thrown a bunch of obsolete units into a city just to slow down his attacking army until my good units could get there and save the city. Of course, that also gives him exp, but sometimes those few extra turns for his infantry to kill off your archers or whatever can save your city.

[Velociryx]

d'oh! I stand corrected on the cultural defense thing, but I guess I'll have to agree to disagree with Wodan re: disbanding troops.

As I read the initial question, I didn't "get" that we were necessarily talking about an army only the size up to the "no maint" limits...since it's pretty easy to exceed those limits most of the time, I did not realize this was a constraint (I'm not sure it would alter my thinking much in any case).

I'd also say that of the two options mentioned (disband or upgrade--or the default choice...nothing), I'd still only rarely opt to disband, since there are so many other things I'd rather do with those hammers (disbanding, for me, is the choice of last resort, as it essentially wastes the hammers spent....if I do ANYTHING else with them, at least I'm getting something for those hammers, right?)

Re: conquest with outdated units....it's been my experience that it's not too hard to take at least one city with an "outdated" division....post construction (which covers 99% of the upgrade situations we're talking about--warrior to spear/axe being the notable exception), with a fistful of cats to do the heavy lifting, warriors and archers can stroll in and take a city with little trouble--granted, you'll not be wanting to pick the city held by scads of modern (and by this I mean, "later era" defenders, not "modern era" defenders) units, but again, the right tool for the job...you'll not naturally gravitate toward sending your older troops to such a city in any case, right?)

So I'd have to say I stand by my earlier comments. For me (and with the understanding that your own mileage may vary), disbanding would be my option of last resort....*especially* if I'm not significantly over the "freebie" limit....it's cheap enough to keep 'em around and use 'em for something.

$0.02

-=Vel=-

[MyOlde]

Sorry, but what are "cats", above?

[Velociryx]

Catapults...and my apologies for the unexplained shorthand.

-=Vel=-

[Blake]

Upgrading is really simple;

Upgrade your elite units, generally those which will be "optimal for the era", meaning they have 3 relevant promotions...
Like upgrading a City Raider 3 Axe to a Mace =
While upgrading a Combat I+II + Shock Knight to a Cavalry = , because the shock promotion isn't relevant...

So upgrade the elites, and just use the rest to make up numbers. Numbers are always useful.

Also, if it's upkeeps you're worried about, just disband the oldest units with the poorest promotions. Generally speaking beyond the middle game, units with only 1 promotion should be tossed out especially when you can now train units with more exp thanks to civics or settled GG's. Of course it's best to toss them out by tossing them at the enemy but don't be too shy about disbanding rubbish units which are truly serving no purpose...


I tend to use the rule that, if a poor unit will sit around soaking up upkeep (no other purpose) for more turns than 1.5x it's hammer cost, then it should be trashed, because that lost gold would have been better spent on research or even upgrading a real unit.
But making this decision requires some forecasting skills. Like if you needed quite a large army to trash a neighbor, and then only had one neighbor left, who is friendly and a good trade partner, you might decide it's better to disband some of the poorer units, especially if you can replace them from a GG-fired military pump.