While answering a question in the discussion forum I started looking into how to make the most out of my empty nodes. As I discovered how the cost scaling works and reached a conclusion regarding how I should act I thought It'd be useful as a guide to share!
The mentioned thread: https://steamproxy-script.pipiskins.com/app/2120250/discussions/0/4358996025341775119/

Get ancient asap for lumber and to start wild growths.

Three simple paths for how to treat an empty node for long-term cost- and node efficiency:
Path 1: Construct a passive building. (Gold mine/ Training facility) (Probably build Training facilities only, check part 4 for a more in-depth consideration.)
Path 2: Construct a one-time resource building (Lab/lumber) then switch it out with a passive bonus building (probably training facility. Check part 4). Remember to get the node to lvl 5 before you switch away from the one-time resource building.


Choose your preferred path based on if your current goals demands one-time resources or not!

Exception if you're rushing DNA for important evolutions (or rushing lumber mills). In that case you can build-and-destroy labs on the same level 5 node. Preferably spread this strategy out among several nodes if possible but if you're early game your options may be limited and you'll have to stick with one or a few nodes.

Keep in mind that constructing a building increases the cost of that particular building in all worlds. you can fight this for constructed training facilities and gold mines by replacing them with other buildings.

This was the original question I answered in a discussion thread, and yup! Every time you construct a lumber mill or a laboratory on a maxxed node you gain the resources you would've gotten had the node been a lumber mill/lab when you captured it, then levelled it up to max stars.

That means that if you have an empty node at lvl 5, then construct a lumber mill, you will get the lumber of a lvl 1+2+3+4+5 lumber mill (including wood multipliers from the ancient legendary tortoise, if it's active).

If you then construct a lab on the same node then you will keep your lumber (minus whatever lumber the lab costs ofcourse) and gain lvl 1+2+3+4+5 worth of DNA.

If you then rebuild another lumber mill at the same spot, you'll gain another "batch" of lumber.

It's been confirmed in this thread that the passive bonus from training facilities disappear once you destroy it: https://steamproxy-script.pipiskins.com/app/2120250/discussions/0/4358996025341775119/

We can assume this is also valid for Gold mines, because their scaling behaves similarly which i will go more into later.

To clarify before we start, with "passive" facilities I refer to gold mines and training facilities, and with "one-time resource" facilities I refer to labs and lumber mills. Also I refer to these constructed buildings as both "facilities" and "Buildings". So when I say "facilities" I don't mean "training facilities" unless it is specified.

All constructed facilities scale both with the number of facilities constructed before on the current node AND in relation to what facilities of the same type have been constructed before in any world. The latter of these differs a bit between passive and one-time resource facilities.


All facilities scale the same way according to how many facilities of any type have been constructed on the selected node before, and the formula is easy:

If it's the first time a building is being constructed on the node, the price remains unchanged at a (1x) multiplier. If it's the 2nd building then it's a (2x) multiplier, and if it's the 3rd building then it's a (3x) multiplier.


There are however two types of cost scaling in relation to previously constructed facilities of the same type:
Type 1: Passive buildings (Goldmines and Training facilities)
Type 2: One-time resource buildings (lumber mills and labs)


The scaling types differs as such:

Type 1: Passive buildings:
Every currently active facility of the same type that you have constructed (not discovered in the map) increases the base price of future facilities. For goldmine the increase is by 50 lumber and for training facilities it's 100 lumber. This makes sense because you lose these passive bonuses as you demolish the buildings, and hence the base price decreases with it.

Type 2: One-time resource buildings:
For every type of this building constructed before, either on this node or on other nodes, active or demolished, the base price increases. 5 gems increase for lumber mills and 150 lumber increase for Labs. That means if you've built and demolished a laboratory on ONE NODE ONLY 100 times (not that I think you'll ever have to do this), then the base cost of building another laboratory on any node has increased by 100 x 150 = 15000 lumber. Even if you have no other constructed labs on the map at all. This makes sense as you keep the facility's resources even when you demolish the building.


So the formula for the wood/war gem price of facilities looks like this:

(Initial building cost + type 1 or 2 scaling increases) x (1+How many times a building has been constructed on this node before)

So if we say the first lumber mill costs 5 war gems, that you've built a total of 10 lumber mills in the world before and you've built 2 (of any) buildings before on the current node, the price of a new lumber mill would be this:

(5 + 5x10)x(1+2)= 55x3=165 war gems.

Sadly I'm unaware of what the initial cost of the different facilities are, but they are quickly dwarfed by the base price increase from building new facilities of the same type.

It's nice to know, as implied from the information from this part, that it's hard to ruin costs for later by screwing up in ignorance early game. For even if you screwed up maximally and built/destroyed a lot of different facilities on lvl 1 nodes then it won't impact the cost of passive structures on other nodes, and the high cost multipliers on the current nodes won't transfer to other nodes either. The only bad consequence will be the relatively low flat base cost increase on new labs and lumber mills. Not to mention your gems income will increase later in the game allowing for more resources for lumber mill constructions!

Also right now Ancient gives x100 lumber instead of x10. If the devs don't change this then you've got access to plenty of lumber to experiment or screw up with.

Justification for getting the legendary tortoise "Ancient" as early as possible:

1. Massive lumber income increase which is PRACTICALLY NECESSARY especially for early-game. (and therefore important for this guide)
2. Getting wild growths started will be less annoying for later, so you don't have to camp at every maxxed node for 10 minutes. Update: After I have mastered both legendary tortoises and given both of them a few weeks of being active, Gorelock seems to just absolutely outdo Ancient in terms of damage increase, with the exception of several units whose entire squad rarely seem to fully die (mouse rangers and sparrow squadron in particular). Therefore wild growths can't really compete with Gorelock, and is more of a compensation that you get while the additional lumber is what is truly useful.


The competing legendary tortoises early-game after Ancient would probably be Gorelock and Rindolf.

Rindolf: Most important trait imo is the winter miner granting tortoise keys and mutation keys. I assume these synergize with miners first evolution that grants up to +10% chance to find rare currency. While this is a good upgrade, equipment from tortoise keys quickly gets replaced by more powerful equipment and the mutations get more powerful as you progress in the game. Not to mention their drop rates are very low unless you heavily invest DNA into this evolution over other argueably more useful evolutions. It's a nice bonus to have but his usefulness pales in comparison with Ancient and Gorelock.

Gorelock: 2nd passive possibly gives all of your squads (up to) 10x damage permanently for every generation you complete. Stacking multiplicatively. It's an incredibly good buff, but over time is probably overshadowed by Ancient's wild growth.


My suggestion as to how to progress with tortoises:

Once you get a tortoise to level 3000 you get both their passives permanently. This also requires you to invest an additional 200 gems to increase their veterancy level.

This is a slow process and not one to be done purely manually unless YOU ARE INSANE!!!

Pack Mouse's last evolution, "collecting gear", automatically collects tortoise food when upgraded to level 2. (Even though it doesnt state so). Additionally, if you wanna go the extra mile, you can get the adventure mode upgrade that gives +25% tortoise food spawn rate.

Ancient is the most powerful early game, so I'd recommend starting with him. Once you get him to veterancy level 3 level 3000 (or get a decent batch of lumber) then go over to Gorelock. Once you've maxed Gorelock then you're a lot more free regarding who to pick. If you still want to think long-term instead of just jumping to your next favourite legendary War Tortoise then getting Navy (to level 1000 only) seems to be a good idea for the valueable currency. Rindolf (to level 1000 again) after that for the same reason then Crusader ( to lvl 3000) for the XP boost.

I will now try to discuss how we should go about choosing what passive buildings we should use.

First, let's look at the passive building options and their buffs at max capture level:


Gold Mine: +250% away money. Note that this only benefits you when you aren't playing the game.
I believe this bonus is multiplicative because the training facility is multiplicative, they both scale similarly, and based on the screenshot I took after coming back to my lvl 1 tortoise in a new generation:

As you can see, the "away money" i earned while afk is 630 cash, while I earned about 30 million "extra cash", which I assume are primarily due to gold mines. Needless to say, this number wouldn't be reachable if gold mines didn't stack multiplicatively.


Training facility: x1.75 DPS
Confirmed by devs to stack multiplicatively. Note that this only benefits you while you're playing the game.
Do note that enemy HP increases by x1.55 per level, meaning that a training facility will "only" provide slightly more than an extra level in pushing power. More accurately, you get a bit more than 5 levels worth of extra damage (or pushing power) if you build 4 training facilities.


I can imagine three types of paths as to how we distribute passive buildings on empty nodes. I will also include estimated numbers as to their effects If we were to distribute them across 20 hypothetically maxxed nodes. I will also include a discussion/evaluation of each path. The order is from 1st which is the cheapest one Lumber-wise to 3rd which is the most expensive.

Path 1: Building a balanced amount of gold mines and training facilities. The most likely ones here are either 2 gold mines for every training facility, so that their wood price stays about the same, or building them at a 1:1 ratio.

2:1 ratio -> 13 gold mines and 7 training facilities -> 3 500 000x offline income and +9 levels of pushing power.

1:1 ratio -> 10 gold mines and 10 training facilities -> 100 000x offline income and +12 levels of pushing power.

The balanced one is the cheapest one. It gives you both active pushing power and more away money. It's important to note that if you can push your run further you'll also earn more money! With this path you'll both get a nice cash boost when you come back from offline while also getting a few extra levels of pushing power on all of your runs.


Path 2: Solely (or mostly) building gold mines. Considering how every new gold mine increases the total amount of generated offline money by +250% then gold mines stack excellently and can become a terrifying amount of money fast.

20 gold mines -> 10 000 000 000x offline money.

So as we can see, the gold mines seems to scale quite well. Especially if you re-invest that money into Mouse Miners to the point where they can get you that amount over and over again consistently.
20 gold mines would be enough to turn an "aa" amount of money into a "ad" amount of money. Which is quite significant, but from my experience it's not enough to make a very big difference. And it does becomes weaker if you aren't near your max level, while their usefulness disappears completely in a generation where you don't turn the game off.
Sadly I do not know how price scaling works. More precisely, I don't know how much more money you'd need in order to afford to push one more level, or how income scales. If anyone could come up with a number regarding this then that'd be great! But it seems like gold mines only are only somewhat useful, and that's when timed properly.


Path 3: Solely ( or mostly) building training facilities.

20 training facilities -> +25 levels of pushing power.

Lastly, we're stuck with only more pushing power. This extra pushing power will be active in every run you do and it does also increase your money gained as you earn more money at higher levels. It's worth noting that it's usefulness isn't dependent on how you time your offline progress.


Conclusion:
"Gold mines only" (probably) has the biggest potential when it comes to how much extra DPS you can get, but that requires carefully timing things well. In contrast by going with training facilities only you'll always benefit maximally from the facilities, even if you don't go offline at all, but you can't expect to come back to the game with a decent cash boost. As for the balanced path you'll get a nice portion of both worlds. It's important to note that none of these passive buildings notably increases your survivability past the early game where you can summon more units to draw aggro. So even if you get +25 levels worth of damage or 10 billion the times of offline money you may be destroyed by an unfortunate bullet barrage long before reaching the potential of this extra pushing power.

Personally I'll go with path 3, Training facilities mostly, because I don't want my progress to be dependent on when I go offline and I don't see gold mines as useful.

Any other opinions regarding this would be much appreciated!

Read the tl;dr part at the start of this guide :)

Justification for the two simple paths mentioned there:
1. Every maxxed empty node will become permanently useful by adopting a passive building.
2. Costs will remain about as low as you can get them while still allowing for the construction for one-time resources from labs and lumber mills.

Note that if you wanna build more lumber mills and labs for the one-time resources, then demolishing/rebuilding on several maxxed nodes saves costs compared to just building/destroying on a single node.


Thanks for reading, and I hope this was useful to you!