When playing a game like They Are Billions it helps to know the relative advantage and cost of each troop and each building. How quickly will upgrading a tent to a cottage take to pay off? Cuz surely it’s a profitable upgrade, but if I do it just two days before the final wave, will I actually get money out of it?
Should be easy! Let’s just take all the units and shove it in a spreadsheet and convert everything to gold. But… there’s a problem.
Everything doesn’t easily convert to gold. A tent takes gold, food, and energy and provides two workers plus 8 gold per tick (a tick is 8 hours in game). Food requires gold and energy, but also consumes gold every tick. Energy requires gold and workers, plus gold every tick. The net result of this is that we can convert everything to gold, but we’ll also have gold per tick (which I’ll call g/t herein) in our evaluations.
I decided to work around that by converting a tent to a strictly gold-and-g/t cost; since the g/t value was positive, that means a tent is a way to invest gold and get the gold back every tick. This is a bit tricky because, how do you account for workers? Tents give you both workers and gold income. There’s no way to get workers without also housing them (except for the Inn, which is not a normal structure), and the only other way to get gold income is to put a quarry next to a gold deposit — which isn’t a normal, easy thing to do. I could consider workers free; but they’re not. If you need to put up a building (say, a ballista) you need workers. You gotta outlay some cash to get that worker. If you bake in the tax money you get from the worker, and then convert that to static cost, the net static cost of the worker becomes zero — which is what I ruled out before (i.e. considering workers to be free).
The game is not zero-sum. In order to build pretty much anything, kinda the first thing you need to do is to build a tent so that you have workers — and that means you get income. Need more energy? Build a tent, then a mill. Want a ballista? Build a tent, then a mill. For the most part, there’s not a lot that actually costs money. Food does — because tehy don’t require workers. You have to pay for the energy and the materials used, but the only worker you have to hire is the sliver of a worker to provide the energy you need.
I decided to just let the numbers be. The one exception is converting gold per tick to an up-front cost; I ignored the worker that’s the byproduct of this exercise and instead just looked at the upfront cost.
Defensive Structures
So, how many snipers is a ballista worth? About 1 and 2/3rds.
There are, of course, complications here, too. Snipers and ballistas are both overkill for small hordes of runners, Aged Infected, and even executives. Both shoot fairly slow, and although the ballista can kill three at a time, the sniper is mobile. The sniper is about 50% more cost effective when killing a single Infected, since the sniper is overall cheaper. Against a swarm of weenies, the ballista (due to its area-of-effect attack) is about twice as cost effective. Against chubbies, they’re equally cost effective. And against the end-game swarm of Billions — which is mostly executives, colonists, and special infected — the ballista again pulls ahead to twice as cost effective.
But what about hitpoints? Ballista wins — unless you put the snipers in a stone tower, which allows them to soak 500 points of damage each (ie 2000 hp total) and then run away and continue to be used to defend the next line of defense (assuming you have one).
Yeah, ok, this is all basically what we knew before. Except there are some keys here — except for mobility, the ballista is more cost effective. If you want to defend a chokepoint against a wave, either the final swarm or an earlier one, then the ballista is probably a better choice, especially given the sniper’s extremely slow movement speed. The ballista will clear the wave faster than 1.67 snipers will. Conclusion 1: Yes: static defenses are worth having. Provided they’ll be used.
What about ballista versus executors? This gets tricky due to the extreme cost of the executor; it’s very close to three times the cost of a ballista (1593g versus 544g, once you convert the wood, the iron, the energy, and the workers to a fixed gold cost). Would you rather have three ballistas, or one executor? One major advantage is the executor only takes up four tiles, whereas you need three times the space for the ballistas. Ballistas have an advantage in that, if one goes down, the other two are still up. However, looking at their DPS numbers, the Executor pulls ahead in cost-effectiveness against swarms — which is what you want them for. You’d need more than five Ballista to match the damage output of one Executor. In terms of making stuff dead, Conclusion 2: Yes, the Executor upgrade is worth it. (The 20 wood that you ‘lose’ by upgrading instead of building an Executor directly is only about 4g in value; not worth it except in the opportunity cost of the 20 wood — are you desperate to get some mills or power plants placed at the same time?)
Basic Units
One thing to note is that veteran units are roughly twice the value of normal units; except for the archer, which in addition to firing twice as fast, needs between a quarter and a third fewer shots to kill smaller zombies due to its bump from 10 damage per shot to 12 damage. Weak infected, with just 35 hit points, die in three shots instead of four; and runners, with 45 hit points, die in four shots instead of five.
Soldiers cost about three times as much as rangers, and they do three times as much damage. Hence, against individual zombies (especially the weak ones or the non-exec runners) these two units are equal in value, except for the rangers longer range and faster running speed. Their cost-effectiveness remains remarkably similar no matter what they face. But again, this is cost-effectiveness. I often use 10 soldiers and 10 snipers to clear VODs; if I was using rangers instead of soldiers, I’d need to send 30 rangers. The Veteran Ranger’s extra damage bump makes them 10-50% more cost-effective than Veteran Soldiers. So go ahead and recruit rangers in the early game; once they’re veterans, they’ll kick ass. But if you’re not getting them levelled up, it’s probably better to use Soldiers and Snipers.
Snipers are about 30% more expensive than soldiers. They have a very slow fire rate but high damage; against weenies, they’re not cost-effective. Against executives, buildings, and chubbies, however, they pull ahead. Against harpies and venoms, though, the target’s 120 hp doesn’t mesh well with the snipers’ 100 point of damage; most of the sniper’s second shot is wasted. The advantage snipers have here is extra hitpoints (to survive attacks) as well as a greater range, meaning they can take out venoms while they’re still at range (for example, on the far side of a defense). Soldiers wouldn’t be able to fire that far, and Rangers would take a long time to eventually kill the venoms — probably dying before the Venoms do. This puts Snipers in a unique position: Conclusion 3: Snipers are great against buildings and chubbies, but overkill against everything else, except for the special case of venoms on the far side of walls.
End-game Units
Lucifers have been buffed recently, and due to their high rate of attack, area-of-effect, and damage-over-time, they are by far the most cost-effective troop for handling waves of weenies and the lesser runners. They’re still way more cost-effective than any normal troop against a swarm of executives, and only slightly beaten out by Titans (and beaten quite a bit by Executors). They can handle harpies or chubbies, but not as effectively as lesser troops, and their range means they have to charge Venoms in order to kill them. I pretty much only use them on the final two waves (and that’s mostly on the last wave), where their advantage over an Executor is that they can move. If you put one in the wrong place, that can be fixed!
Thanatos, which costs just 3% more than a Lucifer (that’s the 10 extra iron), are less cost-effective, but the exact math depends on how many troops they hit. One of the primary differences here is that the Lucifer sends out his burst of flame twice a second, while a Thanatos only fires a missile once every three seconds. The only real advantage that a Thanatos has is range; it can hide out behind walls and run away to fight another day, whereas a beleaguered Lucifer is going down. Conclusion? I don’t really have a clear result here. I use a mix of the two troops. Although the Lucifers often die in that final wave, it is the final wave and it’s not like keeping them alive for the next fight is of any use. There’ve been some really close battles where I’ve been able to shuffle Thanatos around, and that’s helped, but I still like having a few Lucifers to reduce the clutter at my most-sensitive chokepoints.
That brings us to Titans. They are by far the most expensive unit in the game; 34 times as expensive as a ranger, and twice as expensive as an Executor. They don’t even have the highest DPS in the game — that would be an executor. Against swarms, they are the most cost-effective troop in the game, but again less effective than an executor. Their advantage is free healing and mobility. Built an executor in the wrong place? 1600g down the drain. Titan in the wrong place? They move fairly fast. So let’s say you have two Ballista ($544 each). Do you upgrade one ($1050), both, or recruit a Titan ($2960)? Upgrading both is cheaper than recruiting a Titan. They’re not competitive to normal troops against buildings (or Chubbies or Giants), but they are tremendously useful against hordes of Infected, such as those that get released from VODs when you poke them. And if you use one to clear a VOD, you can then use it (again) against another VOD, and a third VOD, and at multiple chokepoints, and against the final Swarm.
So when and/or what do you use Titans for? When you don’t know if you need 2 or 3 Executors, or if you don’t know which of two nearby chokepoints will be hit harder, or if you’re expecting one chokepoint to be overrun but you have a second one to fall back to. One of the most common points I’ve been making is that mobility is one of the most distinguishing characteristics of many choices, and that’s the case here. Conlusion 4: Use a Titan instead of two Executors when mobility is important.
Economy Buildings
Finally, let’s look at the various economy buildings. There’s two groups to consider here: the upgrades (for Housing or to the Advanced buildings), and the options (for Food and Energy).
Generally, the advanced upgrades are not cost-effective. They’re all uniformly bad; roughly 3 or 4 times the cost of a basic unit for only twice the production. In some cases, less than that, as Mayors that buff Farms or Quarries or Mills don’t also buff the Advanced versions — which could mean a 67% boost in output for 300% more dough. There’s only two rough categories of reasons to choose to build the Advanced versions: (1) you’re out of space, or (2) a quarry is perfectly placed in an awesome location.
The housing upgrades are all very very valuable. Do them as soon as you can! Constructing a tent takes just 4 ticks to pay off (that’s 1.3 days). After those 4 ticks, the tent’s income is gravy. Cottages take 7 ticks to pay off, and Stone Houses take 8 ticks. Let’s say you’ve got some food, and are considering whether to upgrade a Tent to a Cottage, or if you should just build a new Tent. Wood costs roughly 1.5g each, so the upgrade costs 91.8g and you get 10g/tick from it; building a new tent costs just 30g and you get 8g/tick. It will take 31 ticks (just over 10 days) for the cottage upgrade to pay off; with a bank, this drops to 21 ticks (7 days). Upgrading a cottage to a house, instead of building two more tents, will take 23 ticks (just under 8 days) to pay off, or just 4 days with a bank. It seems that building cottages early will set you back pretty hard, yet those first 20 days are the easiest. I’m going to think about this some more and throw it into a separate post.
In summary, don’t build a tent (in a 100-day game) after day 92:07 (day 92, seven hours), don’t upgrade a cottage after day 89:15, and don’t upgrade a tent after day 86:15. This is assuming you don’t spend any money once the wave enters the map, which isn’t exactly true — but the general idea is that I spend money queuing up troops and buildings before the wave hits, and if I haven’t started construction before then I generally don’t afterwards. Sometimes, yes, I’ll throw up some extra walls or try to build an extra ballista once I see the final waves — but usually it takes a day and a half to clear the wave and it makes no difference if I’m earning an extra 3g per tick or not.
Food. If you can get 10 or more food from a Hunter’s Cottage, it’s a better option than a farm. I tend to place them anywhere I get 8 or more, as food is always needed and, often, I run out of easy places to put farms. A fisherman’s hut that produces 11 food is about equal to a farm, and it’s easy to find a place to get more than 11 food. Huts cost the same as Cottages, so if you’d place a Hunter for 8 food, you should place a Fisherman for 8. Farms are … farms. You use them because you absolutely need massive amounts of food. Winning requires 1000 to 2000 citizens, maybe even more at the highest difficulty levels, and to get there you’re gonna need farms. Also note that the Crystal Palace is roughly competitive with the other food sources; just note, however, that it takes 7-9 ticks (which is 2-3 days) for housing upgrades to pay off, so you’re going to want to complete all of your housing upgrades three days before the final wave hits — and even then, just because you’ve scraped a bit more income out of the world doesn’t mean squat if you don’t spend it. And if you lose the game with 700 food that you never spent? You coulda bought some Titans, or upgraded a dozen (!! a dozen !!) ballistas.
Energy. There’s four options here: the mill, advanced mill, power plant, and the Wonder. I’ve already addressed the Advanced Mill: don’t do it, unless you’re very tight on space. The Power Plant is 6.3x times as expensive as the mill but only provides 5.3x the power. It’s worth it primarily for the space savings; damn do I hate mills and their restrictions! Like the food Wonder, the Lightning Spire is roughly cost-effective with the Mill and the Power Plant. Likewise, it’s effective if you’ve got a lot of extra cash and can afford to have hundreds of energy being produced that you’re not using, for days and days. Another thing to note is that if you’ve got some far outlying Mills or Power Plants, they get infected, and your power goes negative, every single one of your static defenses will stop working, even if you only have a 1 unit energy shortage. If you’re expecting your outer buildings to be corrupted, it’s best to have a core of energy close to your Command Center, or a surprising excess of power.
Final Word
One thing I haven’t addressed is the shocking tower. This thing can be incredibly effective against the final waves, but it takes hella juice. I haven’t used it enough to get a feel for what it’s equivalent to, but, roughly, it feels like it should be the equal of a couple end-game units. I might look into it later.
Also, check out my YouTube channel — Pixel Wight. Most of these posts are up there as well, usually with better graphics!