• 📚 A new project from the admin: Check out PictureBooks.io, an AI storyteller that lets you create personalized picture books for kids in seconds. Give it a try and let me know what you think!

Uprgade vs. Build New

JDAllison

Prophet of Doom
Joined
Sep 5, 2007
Messages
123
Location
Deep in SabanNation
Howdy.

I apologize up front if this has already been dealt with. I ran several searches and found nothing.

I've been doing some researching and some thinking, and am now ready to post my question/theory. In my current, regent-level game, I'm playing as the Romans. Since I'm trying to win by conquest, I've been building far more units than I usually do (I'm an expantionistic builder, I guess). I don't own Sun Tzu, but am about to take the city that does. I do own Leo's.

Over the course of the past 4700 game-play years, I've managed to amass a pretty good number of obsolescent units. Units that are past their prime, but reasonably ok for defense and peacekeeping. Since they're cheap, I also tend to build a ton of warriors early-on. I'm up to musketmen right now, but still have the odd warrior, spearman, legion, and pikeman hanging around.

My theory is this:

As a general rule is it better to upgrade a unit, or is it better to scrap him and start over? I'm assuming that you'd use what shields he gives as a starter on the new unit, i.e., disband him in the city building the new, modern unit.

If I disband, say, 15 warriors in Rome, I get two shields each. That works out to be 30 shields -- enough to build a Legionary or, later on, a Pikeman. I get a new unit, costing 1gpt in exchange for 15 units at a total cost of 15gpt. However, I am losing a net of 14 units that could be keeping barbs or "friendly" AI troops out of my cities. Then there's the lost shields issue. It costs 10 shields to build a warrior, while disbanding him nets only 2 shields. I've lost 8 shields in the bargain.

For instance, I might have 15 cities, each guarded by one warrior apiece. I can call all 15 together into one city, disband them all, and get a new Pikeman. That leaves me with 14 undefended cities, however, as well as 120 lost shields. (I know that a lot of players leave cities undefended, but I've had my butt handed to me too many times for me to try it again any time soon.)

On the flip side, I can build a barracks in several cities, which I need anyhow to produce vet units and repair injured ones quickly. This allows me to gather all 15 warriors and upgrade them to MDIs for 90g each, or 1350g. Meanwhile, my cities are building additional units or improvements (like libraries or marketplaces, etc). By disbanding all 15 and building one unit, I am forcing my other 14 cities to build a unit apiece, which can take a while in high-corruption cities. By upgrading I am saving time. At some point, however, it becomes more expensive to upgrade than to disband and start over. For instance, in the late game, when my core cities are churning out anywhere from 25 to 45 shields per turn or more, a mech infantry costs only about 2-3 turns, where an upgrade will cost a pretty good bit. Upgrading a spear to a MI is something like 360g (I think- I don't currently have a saved game with any unpromoted spears. If anyone knows better, enlighten me.) At any rate, it is a decent chunk of change.

Then factor in the multiple upgrades. From spear to pike to musket to rifle to infantry to MI. That's 5 upgrades at 30g each, or 150g over time. While that is cheaper than a rush-build, it is 150g that didn't get used in research or science or in rushing a city improvement.

Now, add in the Leo's effect, where all upgrades are half price. Now, it clearly becomes cheaper to upgrade than to disband and rebuild.

My thesis is this: As a militaristic civ looking to dominate by conquest, I need all of the units I can get. Especially early on, upgrading is superior to disband/rebuild. As a builder civ, with around 90 cities (which I regularly hit on huge maps and fewer than 6 civs), and about a unit to a unit-and-a-half per city, upgrading is not quite so cheap and less attractive. With only a few, well-developed cities, disbanding and rebuilding is probably the better choice.

Ok, it's open for discussion! In the words of Bruce, "Smite me, Almighty Smiter!" Oh, and the usual disclaimer: "This is my humble opinion brought from my own experience. Opinion may vary, see store for details. Batteries not included." ;) :lol:
 
Upgrading a spear to a MI is something like 360g (I think- I don't currently have a saved game with any unpromoted spears. If anyone knows better, enlighten me.)

In conquests the cost to upgrade a spear to MI will be (110-20)*3=270 gold. In PTW/vanilla it is 180 gold.

I think I'll leave it to someone else to be first to say something about strategy. :confused:
 
As a rule of thumb, if you are playing a science game (which includes 20k) then you desperately want every gp you can get, and you shouldn't waste any on upgrading. If you are playing a military game (which includes 100k) then there should come a point where you stop researching and redirect all your spare commerce into military production. And the most efficient way of doing that, whether or not you have managed to get hold of Leonardo, is to upgrade obsolete units into contemporary ones.
 
the smaller my military the higher portion of it i tend to upgrade. if i am a small nation i expect to be attacked and thus want an up to date army as soon as possible. but if i'm rampaging i have a large army and the obsolete units can be left in conquered cities as military police.
 
even for the science minded one can argue that upgrading a bunch of obsolescent offensive units will allow the quicker conquest of territories that might be used as science farms or as unit support, either way improving the science situation in the long run.

there is also a huge difference between upgrading a 10-shield regular unit to a 110-shield unit (warrior to MI for example) without leonardos, and upgrading a 100-shield vet unit to a 120-shield unit WITH leonardos (tank to MA). the first example says a lot about why i usually disband my warriors.
 
I choose both courses, and sometimes in the same game. If I have a lot of obsolete units hanging around, but I know I'm about to get into a war and I have the cash, a mass upgrade can be just the thing to ensure victory. On the other hand, if I have a mass of obsolete units, am over unit support limits, and need a city improvement or two, I'll disband like crazy.

It just depends on what you need and when you need it.
 
For the record, obsolete is probably a better term. Obsolescent is an adjective to describe something that is becoming obsolete, as opposed to one that already is ;).

Anyway, I tend to do upgrades. It's only when a unit is unneeded do I disband it.
 
Horseman only need to be upgraded to knights and then cav, or to cav directly. After that, you've either won, or you need to build tanks anyway, nut most likely you would have won already.

If you are playing at such a high difficulty, against such powerful enemies, that building a significant amount of defensive units actually becomes worth it, then you will most likely lose most of your old units in combat and you will be constantly building to replace losses anyway.

I don't think there is a choice between science and upgrading. If you want to win with science, the best way to do it on the long run is to conquer fast. So the main focus should be in military to begin with. Once you've conquered enough land, you'd get rid of most of your military, disbanding units for shields to speed-build science improvements in your core.
 
A big factor is probably if the unit is veteran or not. I tend to get rid of regulars and upgrade veterans when I'm upgrading.
 
For myself I usually find it easier to make new units than to save the cash (and stop the research) to upgrade old ones. This is because I usually play with a heavy hand towards science. If I am playing with a unit I know will last for a while (say Immortal or Cavalry rush) then science be damned!
 
I had this problem just the other night, what to do with so many virtually useless troops. I sent them on a suicide mission to discover an AI strength. Was surprized to see them capture a city. In the end they paied for themselves by holding the city. Great because upkeep had been my biggest concern.
 
I only upgrade if I have leftover units from a previous battle.

Also, they would have to be veteran units and I would need to have built Leo's Workshop.
 
Horseman only need to be upgraded to knights and then cav, or to cav directly. After that, you've either won, or you need to build tanks anyway, nut most likely you would have won already.

If you are playing at such a high difficulty, against such powerful enemies, that building a significant amount of defensive units actually becomes worth it, then you will most likely lose most of your old units in combat and you will be constantly building to replace losses anyway.

I don't think there is a choice between science and upgrading. If you want to win with science, the best way to do it on the long run is to conquer fast. So the main focus should be in military to begin with. Once you've conquered enough land, you'd get rid of most of your military, disbanding units for shields to speed-build science improvements in your core.

Well okay, I tend to think more in PtW than C3C. In C3C one needs an ever-expanding empire to run ever-more scientists, but in PtW, 4 turn research can usually be attained with just 1-2 other civs conquered, and further land is of no use other than for the resources it contains. This is often doable with little or no cash spent on upgrading.
 
Back
Top Bottom