SMAX - The Will to Power - mod

[Hagen0]

Nevill is correct. In vanilla, unless you have ample money to rushbuy them immediatly building recycling tanks early actively hurts your development in most cases. Your examples are unrealistic. A new base either has 2 or 3 mineral production. Since early development in WTP is so slow this changes a bit but they still slow down the first colony pod of that base considerably.

[Nexii]

A new base can pay back faster than a tanks. Because the square it works is more than +3 production

And the base square itself is 4 production

So a base should cost more

[Nevill]

I play my RT exactly right. In vanilla my 2-pop bases are on forests if I don't have anything better, so it's 5 minerals vs. 6 minerals, which - as far as ICS is concerned - means exactly 1 turn of difference in the time it takes to build a Colony Pod. Not that big of a deal. But building RecTanks takes 8 turns for 1-1-1, while I could have another base with 2-1-1 and 10 free minerals, which would pop me a former free of support.

It is a very situational facility.

I normally play with pods off, but when I have them on I try to game the odds and put something heavy in the queue.

In the mod... it depends on which version we are talking about? I can talk about the game I just played. With Colony pods costing 60 there is no ICS to speak of. By 2150 I had 5 bases, 3 of them had Rec Tanks. All conditions were as I said - high mineral yield, not enough food, trading extra minerals for a facility that allows faster growth. By 2180 I had 10 bases, and only one extra Rec Tank was added. I had bigger concerns by then. By 2230 I had 25 bases (we played with a few houserules and put a hard limit on the number), and the game was all but won.

So you can either build one colony per 40 turns or build RT and then build them one per 20 turns = it pays itself off in 40 turns and then doubles your production speed for free.

It isn't free, is what I am trying to say. For these minerals you can pop out a Colony Pod and truly double your production by building another base, which would grow and pop more Colony Pods. It's an opportunity cost.

[Alpha Centauri Bear]

Guys, your models are lacking details. Let's review them more closely.

RT is 1-1-1 = roughly 2.5 minerals equivalent. Sure it's pure mineral payoff time is 40 turns but it adds more on a side in terms of nutrients and energy. Nutrients are directly beneficial as they speed up growth = more people and minerals. Plus some energy for research. So it's composited payoff time is about 40 / 2.5 = 16 turns give or take.

Now let's look at colony. N mineral cost (30, 40, 60 - depending on version). It takes one worker out of the producing base. So its negative benefit in addition to colony cost itself. This negative benefit lasts until colony travels to destination. Say 10 turns roughly. With regular 2-1-0 (undeveloped) square its about -3 * 10 = -30 FOP loss. Then you build new base and gets 6.5 mineral worth FOP right there (I count energy = 0.5 min). So the bonus you get now is 6.5 - 3 (pop decrease loss is never goes away) = 3.5. Then payoff time = (N (cost) + 30 (-1 pop loss)) / 3.5 = 17, 20, 26 turns (by colony cost version). Well, it is pretty much obvious that the colony payoff time is no better than RT one. Even for vanilla.

This is a rough estimate. So numbers can vary up and down. But the idea is - you are forgetting population reduction loss in producing base and travel to destination.

[Nevill]

I am not building some theoretical model, I am talking about years of practice. If your model contradicts it, well...

It takes one worker out of the producing base. So its negative benefit in addition to colony cost itself. This negative benefit lasts until colony travels to destination. Say 10 turns roughly. With regular 2-1-0 (undeveloped) square its about -3 * 10 = -30 FOP loss.

A regular square is 1-1-0 (you'd have to luck out for rainy-rolling tiles), and the extra worker eats 2 food it produces. So the raw benefit of an extra pop is 1 mineral per turn, if that.
And if we are talking ICS (aka base spam), the travel times are 5 turns, if that.

So the "loss" is ~5 FOP.

That's just the first thing I noticed. There are also nuances, like how building CP does not always result in loss of pop - if the base is on its last row of food it produces a CP and immediately grows back, losing nothing in the process.


And that's just a CP. The worth of a former that just magically adds FOP by improving existing tiles is even harder to measure. All in all I find that Rec Tanks are a crutch to the expansion, not a necessity.

[Alpha Centauri Bear]

About practice. Have you actually compared two strategies one to one in the same game? If not then it is practical experience all right but not practical proof.

I stand corrected about the population loss impact. Indeed workers need to eat. That was rough estimate only. I can revise it and verify of course.

[Nevill]

Not sure what you mean by the same game, but I've played a lot of multiplayer games. Does that count?

In vanilla, RecTank start is viable for Morgan (who otherwise suffers from low support and lack of minerals) and Lal (who has something to build while researching Formers). The rest are better off expanding.

I've played exactly one WTP game, and ICS is stopped dead in its tracks by an enourmously high CP cost. I've built a few RecTanks just to help pump bases and formers out, but I have outlined those numbers above.

Tayta was the one who did the RecTanks start (being Lal), and he came worse off, though there were a lot of other factors, of course.

[Nexii]

It's only net FOP that matters, remember to factor that in. An extra colonist/worker eats 2 N. So when the base pop goes down you aren't losing 3-4 FOP but only 1-2 FOP

[Alpha Centauri Bear]

Anyway. Forget the math. I believe we have established that higher Colony pod cost decreases appeal for spamming. So would everybody agree to turn colony pod cost back to 6? I think this will slow down initial spamming a bit. But player will be able to catch up later. We can keep former cost low. Does this seem like an acceptable solution?

[Nevill]

6 seems too high. 5 is probably the highest you can go without crippling the early game too much unless you drastically change mineral yields.

Haven't played any games with 4 mineral rows yet.

[Alpha Centauri Bear]

Not sure what you mean by the same game, but I've played a lot of multiplayer games. Does that count?

If one of your human opponents constantly employed RT strategy then you can average how better/worse it is. Otherwise, if everybody resorted to spamming then there is nothing to compare to. It is just guesswork.

Tayta was the one who did the RecTanks start (being Lal), and he came worse off, though there were a lot of other factors, of course.

Yeah. Unfortunately, it is very randomized so one game doesn't prove anything. 😟

[Alpha Centauri Bear]

6 seems too high. 5 is probably the highest you can go without crippling the early game too much unless you drastically change mineral yields.

Haven't played any games with 4 mineral rows yet.

In this case let's leave it at 4, don't be worried about ICS too much and think of other options when they pop up into our heads.

[Alpha Centauri Bear]

Enough about ICS. I have a brilliant idea about how to make support purely proportional so there is no flat component benefiting small bases.

Two options.
I'll calculate total support for all base units together. After all this is what that gets subtracted from mineral intake. Nobody cares about individual unit support breakdown. Keep in mind these are just examples and coefficients may need tuning, of course.

1. Direct cost

total support = #units * SUPPORT cost

SUPPORT = cost/unit
-4   2.00
-3   1.75
-2   1.50
-1   1.25
0   1.00
1   0.75
2   0.50
3   0.25

Since base mineral production is roughly proportional the population, bigger and more developed bases can support more units (SUPPORT rating adjusted).

2. With free tier

Same as above but units up to base size are free support. And, therefore the SUPPORT cost may be harsher. I.e. some slack in beginning but bigger penalty for excessive number of units.

total support = #excessunits * SUPPORT cost
#excessunits = #units in excess of base size

SUPPORT = cost/unit
-4   4.00
-3   3.50
-2   3.00
-1   2.50
0   2.00
1   1.50
2   1.00
3   0.50

[Alpha Centauri Bear]

Oh, and I think Naval Yards I got from the Maritime Control Center fixed my ships in 1 turn. I definitely recall a probe ship at -90% skip a turn and be repaired the next. I thought that wasn't supposed to happen?

# Version 81

* \[fix\] Repair rate in base with corresponding facility configuration setting didn't work.

[Alpha Centauri Bear]

# Version 82

* The Manifold Harmonics is enabled by Sentient Resonance (~80% in research tree).